Natural History of Community-Acquired Methicillin-Resistant Staphylococcus aureus Colonization and Infection in Soldiers

Transcription

1 MAJOR ARTICLE Natural History of Community-Acquired Methicillin-Resistant Staphylococcus aureus Colonization and Infection in Soldiers Michael W. Ellis, 1 Duane R. Hospenthal, 1 David P. Dooley, 1 Paula J. Gray, 2 and Clinton K. Murray 1 Departments of 1 Medicine and 2 Preventive Medicine, Brooke Army Medical Center, Fort Sam Houston, Texas Background. Community-acquired methicillin-resistant Staphylococcus aureus (CA-MRSA) is an emerging pathogen for which the prevalence, risk factors, and natural history are incompletely understood. Methods. In this prospective observational study, we evaluated 812 US Army soldiers to determine the prevalence of and risk factors for CA-MRSA and the changes in rate over time, as well as to determine the clinical significance of CA-MRSA. Demographic data and swab samples from the nares for S. aureus cultures were obtained from participants at the start of their training and 8 10 weeks later. Over this time period, participants were observed prospectively to monitor for soft-tissue infections. S. aureus isolates were characterized by in vitro examination of antibiotic susceptibilities, meca confirmation, pulsed-field gel electrophoresis, and Panton-Valentine leukocidin (PVL) gene testing. Results. At the initial sampling, 24 of the participants (3%) were colonized with CA-MRSA, 9 of whom (38%) developed soft-tissue infections during the study period. In contrast, 229 participants (28%) were colonized with methicillin-susceptible S. aureus (MSSA), 8 (3%) of whom developed clinical infections during the same period (relative risk, 10.7; 95% confidence interval, ; P!.001 ). At follow-up culture, the CA-MRSA rate dropped to 1.6% without eradication efforts. Previous antibiotic use was a risk factor for CA- MRSA at the initial sampling ( P p.03). PVL genes were detected in 66% of 45 recovered CA-MRSA isolates, including all 9 clinical isolates available for analysis. Of hospitalized, 5 of 6 had PVL-positive CA-MRSA infections. Conclusions. CA-MRSA with PVL-positive strains was associated with a significant risk of softtissue infection, suggesting that CA-MRSA may be more virulent than MSSA. Previous antibiotic use may play a role in CA-MRSA. Methicillin-resistant Staphylococcus aureus (MRSA) can no longer be regarded as a purely nosocomial pathogen. Recent reports of community-acquired MRSA (CA- MRSA) infections in patients without identifiable risk factors point to an ongoing epidemiological shift [1 9]. CA-MRSA is an emerging pathogen with an epidemiology and pathogenesis that continue to be defined [10 13]. A recent meta-analysis of studies addressing CA-MRSA in various communities demonstrated a prevalence of 1.3% [14]. In specific pop- Received 26 March 2004; accepted 13 May 2004; electronically published 2 September Reprints or correspondence: Dr. Michael W. Ellis, Infectious Disease Service MCHE-MDI, Brooke Army Medical Center, 3851 Roger Brooke Dr., Fort Sam Houston, TX (Michael.Ellis2@amedd.army.mil). Clinical Infectious Diseases 2004; 39:971 9 This article is in the public domain, and no copyright is claimed /2004/ ulations, hospital admission within the past 12 months has been noted to be a risk factor for CA-MRSA carriage; however, genotypic and phenotypic characteristics of CA-MRSA suggest that most strains have not originated from hospitals [15 19]. Varying definitions of CA-MRSA and the fact that most studies have focused on participants who are already seeking health care in either a clinic or a hospital setting may have limited these prevalence and risk factor studies. The virulence of CA-MRSA may not be limited solely to its resistance to b-lactam antimicrobial drugs. CA- MRSA appears to have a distinct exotoxin genetic armamentarium in particular, the Panton-Valentine leukocidin (PVL) locus which has been associated with severe infections [17, 18, 20 23]. The significance of CA-MRSA and the presence of the PVL locus with regard to virulence have yet to be determined prospectively. Natural History of CA-MRSA in Soldiers CID 2004:39 (1 October) 971

2 The purpose of this prospective observational study was to determine the prevalence of CA-MRSA, to determine risk factors for CA-MRSA, to determine changes in this rate over the study period, and to determine the clinical importance of CA-MRSA by correlating infection rates and the presence of PVL. MATERIALS AND METHODS Study design. We conducted a prospective observational study assessing CA-MRSA epidemiology in US Army soldiers. Demographic data and swab samples from the nares for S. aureus cultures (hereafter, nares cultures ) were obtained from soldiers upon arrival at Fort Sam Houston, Texas, for training and again 8 10 weeks later. Clinical presentations of infections were correlated with nares culture results. S. aureus isolates were characterized by examination of antibiotic susceptibilities, meca confirmation, genotyping, and PVL gene testing. The Brooke Army Medical Center Investigational Review Board approved the protocol. Study participants. US Army personnel enrolled in the Health Care Specialist Course from 25 August to 18 December 2003 were eligible for the study. This course trains soldiers as combat medics, with the first 10 weeks being entirely outside a health care environment. These soldiers were communally housed in barracks under a strictly controlled training schedule. Specimen collection, identification, and susceptibility testing. The study was designed to begin on the first day of training. After written informed consent was obtained, a questionnaire was administered to all participants that asked for epidemiologic information involving the preceding year. This questionnaire solicited information regarding participant age, gender, previous training site, previous antibiotic use (and indication), previous nasal or oral corticosteroid use, residence with patients with chronic diseases, residence with a person who had been admitted to the hospital, participant hospital admission, hospital or nursing home employment, and residence with children under the age of 16 years. After questionnaire completion, nares culture specimens were collected with CultureSwabs with Stuart media (BBL). Within 4 h of collection, specimens were plated onto mannitol salt agar. Mannitol fermenting colonies were isolated and plated onto trypticase soy agar with 5% sheep s blood. Reisolated colonies were screened with tube coagulase and Gram staining. S. aureus isolates were tested with oxacillin screening agar (Mueller-Hinton with oxacillin, 6 mg/ml; Becton Dickinson). MRSA isolates were tested for susceptibility by disk diffusion to tetracycline, vancomycin, clindamycin, erythromycin, trimethoprim-sulfamethoxazole, ciprofloxacin, rifampin, linezolid, and daptomycin. Inducible clindamycin resistance was assessed with a double-disk diffusion test (D-test). All microbiological procedures were conducted according to NCCLS guidelines [24]. Eight to 10 weeks after the initial nares culture, a second culture was performed in the same manner. Similarly, a second questionnaire was administered to assess epidemiological information during the interval observation period. This questionnaire inquired about interval antibiotic use (and indication), interval nasal or oral corticosteroid use, participant hospital admission, participant soft-tissue infection, and participant hospital or clinic training exposure. Over the 8 10-week period, participants were observed prospectively to assess for clinical soft-tissue infection. All participants were seen in a single clinic for any acute illness, and when needed, they were referred to Brooke Army Medical Center, a tertiary care center. Health care providers were blinded to the nares culture results. Bacterial wound cultures of clinical specimens were processed in the Brooke Army Medical Center microbiology laboratory. Clinic records were reviewed daily for soft-tissue infections in study participants. Participant record review was facilitated by the fact that they were served by a united health care system with a single computerized records system, microbiology laboratory, and record depository. Finally, a retrospective review of all codes from the International Classification of Diseases, 9th Revision, was conducted to ensure the capture of all participant clinical information. All MRSA isolates obtained from both initial and terminal nares samplings and all MRSA clinical isolates underwent additional laboratory studies. We prospectively defined CA-MRSA as a specimen isolated from a participant without known MRSA risk factors for which the CA-MRSA organism possessed a characteristic CA-MRSA antimicrobial susceptibility profile and a molecular profile different than our prevailing hospital-associated MRSA strain. Our predominant hospital-associated MRSA strain is resistant to all b-lactam agents, fluoroquinolones, erythromycin, and clindamycin. Molecular typing with PFGE. DNA was digested with SmaI and resolved with the CHEF-DRII apparatus (Bio-Rad). Gel findings were resolved by Quantity One specialized software, version 4.1 (Bio-Rad). PFGE patterns were interpreted on the basis of established criteria [25, 26]. Detection of meca. Real-time PCR performed with a LightCycler (Roche) was conducted for amplification and hybridization probe-based detection of the meca amplicon [27]. Fluorescence curves were analyzed with LightCycler software, version Detection of PVL. The methodology for PCR amplification has been described elsewhere [20, 28]. ITBioChem designed oligonucleotide primers for PCR amplification of the luks-pv and the lukf-pv genes. Primer sequences for the PVL genes were as follows: PVL-1 (which encodes luks-pv), 5 - CTGGTGCGATTCATGGTA-3 ; and PVL-2 (which encodes lukf-pv), 5 -CGATATCGTGGTCATCACA-3. S aureus strain ATCC was used as the PVL-positive control. 972 CID 2004:39 (1 October) Ellis et al.

3 Table 1. Baseline characteristics of 812 study participants. Characteristic Value Age, mean years (range) 21.1 (18 44) Sex Male 619 (76) Female 193 (24) Previous training site Fort Benning, GA 150 (18) Fort Leonard Wood, MO 143 (18) Fort Sill, OK 204 (25) Fort Knox, KY 92 (11) Fort Jackson, SC 191 (24) Other training site 6 (1) Home 26 (3) NOTE. Data are no. (%) of, unless otherwise indicated. Percentages that exceed 100% are a result of rounding. Statistical analysis. The null hypothesis was that there is no relationship between the initial sampling results and the subsequent rate of infection. Expecting a prevalence of 3% and an increase of 10% to be clinically significant, we estimated that 800 participants were needed to detect the expected difference in prevalence with a level of confidence of 95% and a power of 80%. The associations of methicillin-susceptible S. aureus (MSSA), CA-MRSA, and no S. aureus with subsequent infections were compared by the x 2 contingency test. Univariate analysis for significant variable associations was performed with the x 2 contingency test. Multivariate analysis for significant variable associations was performed by logistic regression. We used SPSS statistical software, version 11.5 (SPSS), for our analysis. RESULTS Population characteristics. Of 1230 soldiers eligible for the study, there were 812 volunteers, of whom 619 were men. The participants were years old (mean age, 21 years). Almost all participants had arrived directly from the US Army s 5 major basic training sites (table 1). CA-MRSA. At the initial sampling, 24 participants (3%) were colonized with CA-MRSA, and 229 participants (28%) were colonized with MSSA. Seven hundred sixtyone participants were available for the second sampling (94% of the initial 812 participants). Participants had dropped out of the study population for the following reasons: academic failure, disciplinary action, redeployment to another station, and family emergencies. None dropped out because of illness. At the second sampling, the CA-MRSA rate had decreased to 12 (1.6%) of the 761 participants and the MSSA rate had decreased to 152 (20%) of 761. No participant with initial MSSA subsequently developed CA-MRSA. Four (1%) and 40 (8%) of the 523 participants with no initial S. aureus subsequently developed CA-MRSA and MSSA, respectively (table 2). Risk factors. After univariate and multivariate analyses, a statistically significant risk factor for CA-MRSA at the initial sampling was antibiotic use within the previous 6 months ( P p.03) (table 3). Participants who had lived with children younger than 16 years of age were noted to have lower rates of CA-MRSA ( P p.002). Antibiotic use during the study period and a personal history of frequent softtissue infections were significant risk factors for CA-MRSA at the second sampling ( P p.010 ); however, these did not remain significant with multivariate analysis (table 4). Antimicrobial susceptibilities. Antimicrobial susceptibility testing indicated resistance patterns suggestive of CA-MRSA strains in all of the MRSA isolates (table 5). These patterns were different from the predominant hospital-acquired strain at our institution. Eight (18.6%) of the 43 isolates susceptible to clindamycin demonstrated inducible clindamycin resistance. Virulence factors. Of the CA-MRSA 45 isolates available for testing (9 clinical isolates, and 36 nasal isolates), the genes luks-pv and lukf-pv, which signify the presence of PVL, were Table 2. Initial sampling results versus second sampling results. Findings of initial culture (no.) MRSA Findings of second culture, no. (%) of MSSA Staphylococcus aureus not present MRSA (24) 8 (33) a 1 (4) 15 (63) MSSA (214) (52) 103 (48) S. aureus not present (523) 4 (1) 40 (8) 479 (92) NOTE. Data are tabulated from the 761 participants with both initial and terminal samples. Percentages that exceed 100% are a result of rounding. MRSA, methicillinresistant S. aureus; MSSA, methicillin-susceptible S. aureus. a P!.0001, for MRSA carriers versus MSSA carriers. Natural History of CA-MRSA in Soldiers CID 2004:39 (1 October) 973

4 Table 3. Characteristics of groups at the time of collection of the initial sample. Characteristic MRSA-colonized MSSA-colonized Subjects without Staphylococcus aureus No. (%) of 24 (3) 229 (28) 559 (69) Age, mean years SD Sex, no. of Male Female Previous training site Fort Benning, GA 4 (3) 38 (25) 108 (72) Fort Leonard Wood, MO 4 (3) 34 (24) 105 (73) Fort Sill, OK 2 (1) 83 (41) 119 (58) Fort Knox, KY 6 (6) 23 (25) 63 (67) Fort Jackson, SC 7 (4) 40 (20) 144 (76) Other site 0 (0) 2 (33) 4 (67) Home 1 (4) 9 (34) 16 (62) Antibiotic use a Yes 17 (71) b 98 (43) 245 (44) No 7 (29) 130 (57) 311 (56) Corticosteroid use a Yes 0 16 (7) 40 (7) No 24 (100) 213 (93) 516 (93) Chronic disease c Yes 2 (8) 24 (10) 68 (12) No 22 (92) 204 (90) 488 (88) Lived with health care worker d Yes 7 (29) 97 (42) 190 (34) No 17 (71) 131 (58) 366 (66) Skilled nursing e Yes 4 (17) 30 (13) 54 (10) No 20 (83) 198 (87) 502 (90) Hospitalization f Yes 4 (17) 24 (11) 71 (13) No 20 (83) 204 (90) 485 (87) Worked in hospital g Yes 5 (21) 36 (16) 80 (14) No 19 (79) 192 (84) 476 (86) Lived with children h Yes 6 (25) i 103 (45) 305 (55) No 18 (75) 124 (55) 231 (45) Infection history j Yes 3 (13) 12 (5) 17 (3) No 21 (88) 216 (95) 539 (97) NOTE. Data are no. (%) of, unless otherwise indicated. MRSA, methicillin-resistant S. aureus; MSSA, methicillin-susceptible S. aureus. a Received therapy within 6 months. b P p.03 c Lived with a person with a chronic disease (diabetes mellitus, asthma, chronic obstructive pulmonary disease, dialysis, vascular disease, organ transplant) within 1 year. d Lived with a health care worker or person who was admitted to a hospital within 1 year. e Lived with someone who worked in or lived in a skilled nursing facility within 1 year. f Participant admitted to a hospital within 1 year. g Participant worked in a hospital or skilled nursing facility within 1 year. h Lived with person aged!16 years within 1 year. i j P p.002. Participant has a history of frequent soft-tissue infection.

5 Table 4. sample. Characteristics of groups at time of collection of the second Characteristic MRSA-colonized MSSA-colonized Subjects without Staphylococcus aureus No. (%) of 12 (1.6) 152 (20) 597 (78) Antibiotic use a Yes 7 (58) 28 (18) 129 (22) No 5 (42) 124 (82) 468 (78) Corticosteroid use a Yes 0 11 (7) 58 (10) No 12 (100) 141 (93) 539 (90) Hospitalization b Yes 2 (17) 9 (6) 30 (5) No 10 (83) 143 (94) 567 (95) Infection c Yes 3 (25) 4 (3) 26 (4) No 9 (75) 148 (97) 571 (96) Clinic visit d Yes 5 (42) 3 (2) 24 (4) No 7 (58) 149 (98) 573 (96) Worked in hospital Yes 0 1 (1) 3 (1) No 12 (100) 151 (99) 594 (99) Worked in clinic Yes (1%) No 12 (100) 152 (100) 594 (99) NOTE. Data are no. (%) of, unless otherwise indicated. MRSA, methicillin-resistant S. aureus; MSSA, methicillin-susceptible S. aureus. a Therapy during study interval. b Admitted to hospital during study interval. c Developed soft-tissue infection during study interval. d Treated for soft-tissue infection during study interval. detected in 30 (67%). Of the 9 clinical wound isolates available for analysis, all 9 were PVL positive. Of the 36 nasal isolates, 21 (58%) were PVL positive. Molecular analysis. PFGE of the 45 total CA-MRSA isolates recovered (nasal isolates and clinical isolates) demonstrated 8 distinct genotypes. These strains were different from the predominant nosocomial MRSA strain at our institution. One genotype accounted for 25 (55.6%) of the 45 total CA- MRSA isolates. This predominant strain appeared similar to the pulsed-field type USA300, as recently described by Mc- Dougal et al. [29]. Eight of the 9 clinical CA-MRSA isolates available for analysis belonged to this genotype. Clinical impact. Colonization with CA-MRSA at the initial sampling was a significant risk factor for developing clinical soft-tissue infection during the study period (figure 1). Nine (38%) of 24 participants who were colonized with CA-MRSA developed soft-tissue infections, whereas 8 (3%) of 229 participants who were colonized with MSSA developed soft-tissue infections ( P!.001) (table 6). The infection rate for participants who were colonized with neither CA-MRSA nor MSSA at the initial sampling was 2% (12 participants), which was not significantly different than that for those with MSSA. Of these 9 CA-MRSA colonized participants who developed soft-tissue infections, 8 (89%) had been colonized at the initial sampling with a PVL-positive strain of MRSA. CA-MRSA colonized participants developed the following infections: cellulitis without abscess (1 case), abscesses with no obtained culture specimens (4 cases), and abscesses from which CA-MRSA was cultured (4 cases). The single participant who developed cellulitis had been colonized with a PVL-negative strain of CA-MRSA. The 4 participants whose abscesses were not sampled for culture had undergone incision and drainage. All 4 of these participants had been colonized with a PVLpositive strain of CA-MRSA. The 4 abscesses that were sampled for culture and analyzed all demonstrated the predominant PVL-positive CA-MRSA genotype, and each was indistinguishable from the strain present at their initial sampling. MSSA-colonized participants developed the following infections: cellulitis without abscess (3 cases), abscesses with no obtained culture specimens (3 cases), and abscesses from which Natural History of CA-MRSA in Soldiers CID 2004:39 (1 October) 975

6 Table 5. Community-acquired methicillin-resistant Staphylococcus aureus antibiotic susceptibilities in 45 samples. Antibiotic No. (%) of samples, by susceptibility Susceptible Resistant Intermediate Vancomycin 45 (100) Rifampin 41 (91.1) 4 (8.9) TMP-SMZ 45 (100) Tetracycline 45 (100) Ciprofloxacin 29 (64.4) 7 (15.6) 9 (20.0) Linezolid 45 (100) Clindamycin a 43 (95.6) 2 (4.4) Erythromycin 9 (20) 36 (80) Daptomycin b 45 (100) NOTE. TMP-SMZ, trimethoprim-sulfamethoxazole. a Eight (18.6%) of 43 isolates demonstrated inducible resistance. b Greater than or equal to 16-mm zone considered susceptible, according to the manufacturer. CA-MRSA was cultured (2 cases). Both of these CA-MRSA isolates belonged to the predominant PVL-positive genotype. Three participants with abscesses had them incised and drained, but specimens were not sent for culture. Participants who had not been colonized with S. aureus at the initial sampling developed the following infections: cellulitis without abscess (4 cases), abscesses with no culture samples obtained (3 cases), and abscesses with CA-MRSA obtained on culture (5 cases). Three of these 5 CA-MRSA abscess isolates were available for molecular analysis. All 3 were PVL positive, and 2 belonged to the predominant genotype. Six of the 29 participants who developed soft-tissue infections required hospitalization for parenteral therapy and surgical consultation for the management of abscesses, 5 of whom had PVL-positive CA-MRSA strains implicated in the infection. The remaining participant s abscess drainage was not sent for culture. During the investigational period, 1 patient who belonged to the recruited cohort but had not enrolled in the study developed a CA-MRSA abscess that was complicated by bacteremia. This patient s clinical isolates demonstrated the same predominant PVL-positive genotype found in study participants. DISCUSSION This is the first prospective study of the natural history of CA- MRSA. Overall, this investigation demonstrates that CA-MRSA is a risk factor for subsequent soft-tissue infection and that, likely secondary to unique genes such as PVL, CA- MRSA appears to be more virulent than MSSA. It shows that antibiotic use is a risk factor for developing CA-MRSA. Our study has the benefit of sampling a large group of young, healthy participants from diverse geographical locations completely outside of a health care setting. We found that 38% of participants who were colonized with CA-MRSA at the initial sampling subsequently developed softtissue infections. Moreover, a single strain expressing PVL, a leukocyte-destroying exotoxin that causes tissue necrosis, was responsible for nearly all of the infections. Nasal with MSSA, as well as with MRSA, has already been described as a risk factor for subsequent infections; however, our high attack rate suggests greater virulence beyond b-lactam resistance alone [30 36]. This increased virulence may be explained in part by the activity of PVL. Indeed, PVL has been associated with severe pneumonia and skin infections on several continents [17, 20, 28, 37]. Although PVL is found in only 2% 3% of MSSA isolates, all 9 of our clinical isolates (from abscesses) available for molecular analysis carried the gene [38]. In ad- Figure 1. Flow chart illustrating the initial nasal culture results and subsequent clinical infections. CA-MRSA, community-acquired methicillinresistant Staphylococcus aureus; MSSA, methicillin-susceptible S. aureus. *Panton-Valentine leukocidin (PVL) positive strain. 976 CID 2004:39 (1 October) Ellis et al.

7 Table 6. Initial nares swab culture results and infection rate. Infection during study MRSA-colonized (n p 24) No. (%) of MSSA-colonized (n p 229) Subjects without Staphylococcus aureus (n p 559) Yes 9 (38) a 8 (3) 12 (2) No 15 (63) 221 (97) 547 (98) NOTE. MRSA, methicillin-resistant S. aureus; MSSA, methicillin-susceptible S. aureus. a P!.001, for MRSA carriers versus MSSA carriers (relative risk, 10.7; 95% CI, ). dition, of the 9 CA-MRSA colonized participants who developed infections, 8 of these had been carriers of a PVL-positive strain. Finding PVL in our CA-MRSA strains is consistent with other reports of the same association [17, 20, 28]. Researchers have postulated that CA-MRSA arose from the insertion of the mobile staphylococcal cassette chromosome mec (SCCmec) type IV into a virulent MSSA strain that already possessed PVL and that the presence of PVL along with SCCmec type IV may actually endow CA-MRSA with a selective advantage over other strains [19 21, 39, 40]. CA-MRSA possessing SCCmec IV may even have a faster doubling time that may further augment overall fitness [41]. The initial point prevalence of CA-MRSA in our study was 3%. This finding is similar to that of a recent meta-analysis that looked at CA-MRSA prevalence in various populations [14]. Likewise, the prevalence of MSSA was 28%, which is in keeping with historically reported rates, implying that our sampling technique is valid [30]. The finding that the prevalence of both CA-MRSA and MSSA decreased to 1.6% and 20%, respectively, may be due to increased living space and improved hygiene, compared with the basic training environment to which our participants were previously exposed [42]. No one who was initially colonized with MSSA subsequently developed CA-MRSA. However, this MSSA did not seem to protect them entirely from infection because 5 participants from this group developed CA-MRSA abscesses; the 2 isolates available for analysis were PVL positive. Although risk factors for hospital-associated MRSA have been well described, the same has not yet been done for true CA-MRSA [3, 10, 11, 43]. This may in part be due to disparate definitions of CA-MRSA [14, 42]. In our study, we found antimicrobial use within 6 months preceding initial sampling to be a risk factor for CA-MRSA. Despite the fact that prevalence has been noted to be increasing in pediatric populations, we found that having lived with a child aged 16 years was associated with a lower prevalence of CA- MRSA [44]. An explanation for this finding poses a challenge, because children tend to be more persistently colonized with S. aureus [30]. We suspect that our survey may not have been specific enough to exclude unrecognized factors, and this observation remains an issue for further investigation. It may be that basic training itself is a risk factor: crowding, inadequate hygiene, and possible skin trauma may contribute to S. aureus and thus infection [30, 42, 45]. Outbreaks have been reported in similar environments found in correctional facilities and on sports teams [5, 8, 9]. It would appear that CA-MRSA has some degree of enhanced virulence, compared with MSSA, and that it possesses an epidemiology more closely resembling MSSA than that of hospital-associated MRSA [46]. This would explain the absence of previously recognized nosocomial risk factors associated with CA-MRSA. Given these results, several questions arise. First, what initial antimicrobial therapy should be used for community-acquired soft-tissue infections? CA-MRSA remains susceptible to numerous non b-lactam agents. Our isolates demonstrated this susceptibility pattern, which is consistent with other reports [17]. One of these alternative non b-lactam agents should be considered in communities where CA-MRSA is prevalent, with special consideration given to the presence of inducible clindamycin resistance [47]. Almost one-fifth of our clindamycinsusceptible CA-MRSA isolates actually possessed inducible resistance, which has been noted to contribute to clinical failure [48]. At the same time, routinely submitting incised and drained abscess fluid specimens for culture should also be encouraged because it could not only affect clinical management, but also help define local resistance patterns [47]. Second, it is uncertain whether attempts to eradicate or control CA-MRSA could be effective or required to prevent invasive infection. CA- MRSA decreased in our population during the study period without active eradication efforts, although this decrease did not reach statistical significance ( P p.06). This decreased CA-MRSA rate that paralleled the decrease in MSSA may point to epidemiological similarities between these 2 organisms. In addition, although no participant initially colonized with MSSA subsequently developed CA-MRSA at the terminal sampling, MSSA did not protect these participants from subsequent CA-MRSA infection, which occurred in 5 participants. Natural History of CA-MRSA in Soldiers CID 2004:39 (1 October) 977

8 In conclusion, this study supplements existing knowledge of the natural history of CA-MRSA infection. Our data suggest that CA-MRSA is a significant risk factor for subsequent soft-tissue infection and even bacteremia and that currently circulating CA-MRSA strains may be more virulent than MSSA. This virulence may in part be due to the presence of PVL. These data could help develop strategies for CA-MRSA therapy and control efforts. Acknowledgments We thank Dr. James Jorgensen for technical assistance; Dr. Robert Daum and Dr. Gerard Lina, who provided technical advice for detecting PVL; Sherry Trevino, Alicia Astorga, and Linda Anderson, who contributed to our molecular analysis; Dr. John Ward, who aided us with our statistical analysis; and the Infectious Diseases staff for their support. Financial support. Department of Clinical Investigations, Brooke Army Medical Center. Conflict of interest. All authors: No conflict. References 1. Naimi TS, LeDell KH, Boxrud DJ, et al. Epidemiology and clonality of community-acquired methicillin-resistant Staphylococcus aureus in Minnesota, Clin Infect Dis 2001; 33: Herold BC, Immergluck LC, Maranan MC, et al. Community-acquired methicillin-resistant Staphylococcus aureus in children with no identified predisposing risk. JAMA 1998; 279: Groom AV, Wolsey DH, Naimi TS, et al. Community-acquired methicillin-resistant Staphylococcus aureus in a rural American Indian community. JAMA 2001; 286: Centers for Disease Control and Prevention. Four pediatric deaths from community-acquired methicillin-resistant Staphylococcus aureus Minnesota and North Dakota, MMWR Morb Mortal Wkly Rep 1999; 48: Centers for Disease Control and Prevention. Outbreaks of communityassociated methicillin-resistant Staphylococcus aureus skin infections Los Angeles County, California, MMWR Morb Mortal Wkly Rep 2003; 52: Shahin R, Johnson IL, Jamieson F, et al. Methicillin-resistant Staphylococcus aureus carriage in a child care center following a case of disease. Arch Pediatr Adolesc Med 1999; 153: Suggs AH, Maranan MC, Boyle-Vabra S, Daum RS. Methicillin-reistant and borderline methicillin-resistant asymptomatic Staphylococcus aureus in children without identifiable risk factors. Pediatr Infect Dis J 1999; 18: Lindenmayer JM, Schoenfield S, O Grady R, Carney JK. Methicillinreistant Staphylococcus aureus in a high school wrestling team and the surrounding community. Arch Intern Med 1998; 158: Methicillin-resistant Staphylococcus aureus infections among competitive sports participants Colorado, Indiana, Pennsylvania, and Los Angeles County, MMWR Morb Mortal Wkly Rep 2003; 52: Bukharie HA, Abdelhadi MS, Ibrahim AS, Rubaish AM, Larbi EB. Emergence of methicillin-resistant Staphylococcus aureus as a community pathogen. Diagn Microbiol Infect Dis 2001; 40: Chambers HF. The changing epidemiology of Staphylococcus aureus? Emerg Infect Dis 2001; 7: Moreno F, Crisp C, Jorgensen JH, Patterson JE. Methicillin-resistant Staphylococcus aureus as a community organism. Clin Infect Dis 1995; 21: Gorak EJ, Yamada SM, Brown JD. Community-acquired methicillinresistant Staphylococcus aureus in hospitalized adults and children without known risk factors. Clin Infect Dis 1999; 29: Salgado CD, Farr BM, Calfee DP. Community-acquired methicillinresistant Staphylococcus aureus: a meta-analysis of prevalence and risk factors. Clin Infect Dis 2003; 36: Charlebois ED, Bangsberg DR, Moss NJ, et al. Population-based prevalence of methicillin-resistant Staphylococcus aureus in the urban poor of San Francisco. Clin Infect Dis 2002; 34: Jernigan JA, Pullen AL, Partin C, Jarvis WR. Prevalence of and risk factors for with methicillin-resistant Staphylococcus aureus in an outpatient clinic population. Infect Control Hosp Epidemiol 2003; 24: Naimi TS, LeDell KH, Como-Sabetti K, et al. Comparison of community- and health care associated methicillin-resistant Staphylococcus aureus infection. JAMA 2003; 290: Fey PD, Said-Salim B, Rupp ME, et al. Comparative molecular analysis of community- or hospital-acquired associated methicillin-resistant Staphylococcus aureus. Anitmicrob Agents Chemother 2003; 47: Daum RS, Ito T, Hiramatsu K, et al. A novel methicillin-resistant cassette in community-acquired methicillin-resistant Staphylococcus aureus isolates of diverse genetic background. J Infect Dis 2002; 186: Mongkolrattanothai K, Boyle S, Kahana MD, Daum RS. Severe Staphylococcus aureus infections caused by clonally related communityacquired methicillin-susceptible and methicillin-resistant isolates. Clin Infect Dis 2003; 37: Vandenesch F, Naimi T, Enright MC, et al. Community-acquired methicillin-resistant Staphylococcus aureus carrying Panton-Valentine leukocidin genes: worldwide emergence. Emerg Infect Dis 2003; 9: Boubaker K, Diebold P, Blanc DS, et al. Panton-Valentine leukocidin and staphylococcal skin infections in schoolchildren. Emerg Infect Dis 2004; 10: Dufour P, Gillet Y, Bes M, et al. Community-acquired methicillinresistant Staphylococcus aureus infections in France: emergence of a single clone that produces Panton-Valentine leukocidin. Clin Infect Dis 2002; 35: NCCLS. Performance standards for antimicrobial susceptibility testing. 9th informational supplement. Wayne, PA: NCCLS, Tenover FC, Abreit 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: Bannerman TL, Hancock GA, Tenover FC, Miller JM. Pulsed-field gel electrophoresis as a replacement for bacteriophage typing of Staphylococcus aureus. J Clin Microbiol 1995; 33: Reischl U, Linde H-J, Metz M, Leppmeier B, Lehn N. Rapid identification of methicillin-resistant Staphylococcus aureus and simultaneous species confirmation using real-time fluorescence PCR. J Clin Microbiol 2000; 38: Lina G, Piemont Y, Godail-Gamot F, et al. Involvement of Panton- Valentine leukocidin-producing Staphylococcus aureus in primary skin infections and pneumonia. Clin Infect Dis 1999; 29: 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: Kluytmans J, Van Belkum A, Verbrugh H. Nasal carriage of Staphylococcus aureus: epidemiology, underlying mechanisms, and associated risks. Clin Microbiol Rev 1997; 10: Yu VL, Goetz A, Wagener M, et al. Staphylococcus aureus nasal carriage and infection in patients on hemodialysis. N Engl J Med 1986; 315: Luzar MA, Coles GA, Faller B, et al. Staphylococcus aureus nasal carriage and infection in patients on continuous ambulatory peritoneal dialysis. N Engl J Med 1990; 322: Boelaert JR, Van Landuyt HW, Godard CA, et al. Nasal mupirocin 978 CID 2004:39 (1 October) Ellis et al.

9 ointment decreases the incidence of Staphylococcus aureus bacteremias in haemodialysis patients. Nephrol Dial Transplant 1993; 8: Wenzel RP, Perl TM. The significance of nasal carriage of Staphylococcus aureus and the incidence of postoperative wound infection. J Hosp Infect 1995; 31: Fierobe L, Decre D, Muller C, et al. Methicillin-esistant Staphylococcus aureus as a causative agent of postoperative intra-abdominal infection: relation to nasal. Clin Infect Dis 1999; 29: Muder RR, Brennen C, Wagener MM, et al. Methicillin-resistant staphylococcal and infection in a long-term care facility. Ann Intern Med 1991; 114: Gillet Y, Issartel B, Vanhems P, et al. Association between Staphylococcus aureus strains carrying gene for Panton-Valentine leukocidin and highly lethal necrotising pneumonia in young immunocompetent patients. Lancet 2002; 359: Dinges MM, Orwin PM, Schlievert PM. Exotoxins of Staphylococcus aureus. Clin Microbiol Rev 2000; 13: Wielders CLC, Vriens MR, Brisse S, et al. Evidence for in-vivo transfer of meca DNA between strains of Staphylococcus aureus. Lancet 2001; 357: Baba T, Takeuchi F, Kuroda M, et al. Genome and virulence determinants of high virulence community-acquired MRSA. Lancet 2002; 359: Okuma K, Iwakawa K, Turnidge JD, et al. Dissemination of new methicillin-resistant Staphylococcus aureus clones in the community. J Clin Microbiol 2002; 40: Cookson BD. Methicillin-resistant Staphylococcus aureus in the community: new battlefronts, or are the battles lost? Infect Control Hosp Epidemiol 2000; 21: Harbarth S, Liassine N, Dharan S, Herrault P, Auckenthalr R, Pittet D. Risk factors for persistent carriage of methicillin-resistant Staphylococcus aureus. Clin Infect Dis 2000; 31: Hussain FM, Boyle-Vavra S, Bethel C, Daum RS. Current trends in community-acquired methicillin-resistant Staphylococcus aureus at a tertiary care pediatric facility. Pediatr Infect Dis 2000; 19: Martinez-Lopez LE, Friedl KE, Morre RJ, Kramer TR. A longitudinal study of infections and injuries of Ranger students. Mil Med 1993; 158: De Sousa MA, de Lencastre H. Evolution of sporadic isolates of methicillin-resistant Staphylococcus aureus (MRSA) in hospitals and their similarities to isolates of community-acquired MRSA. J Clin Microbiol 2003; 41: Stevens DL. Community-acquired Staphylococcus aureus infections: increasing virulence and emerging methicillin resistance in the new millennium. Curr Opin Infect Dis 2003; 16: Siberry GK Tekle T, Carroll K, Dick J. Failure of clindamycin treatment of methicillin-resistant Staphylococcus aureus expressing inducible clindamycin resistance in vitro. Clin Infect Dis 2003; 37: Natural History of CA-MRSA in Soldiers CID 2004:39 (1 October) 979