Community-Associated Methicillin-Resistant Staphylococcus aureus as a Cause of Health Care—Associated Infection
- Infectious Diseases Section, Hospital of Saint Raphael and Clinical Professor of Medicine, Yale University School of Medicine, New Haven, Connecticut
- Reprints or correspondence: Dr. John M. Boyce, Infectious Diseases Section, Hospital of Saint Raphael, 1450 Chapel St., New Haven, CT 06511 (JBoyce{at}srhs.org).
Methicillin-resistant Staphylococcus aureus (MRSA) infection was relatively uncommon in the United States before 1975. From 1975 through 1980, MRSA infection spread to hospitals in previously unaffected areas of the United States, and the number of institutions reporting cases of MRSA infection increased considerably [1]. From 1983 through the early 1990s, the prevalence of health care-associated (HA) MRSA infection increased dramatically, and it continued to increase more gradually during the latter half of the 1990s [2, 3]. In 1998, Herold et al. [4] reported an alarming increase in the occurrence of community-associated (CA) MRSA infection among young children with none of the typical risk factors associated with HA MRSA. Within a few years, CA MRSA strains became widely disseminated, accounting for 15%–74% of S. aureus skin and soft-tissue infections seen in emergency departments [5–8]. A recent population-based study of invasive MRSA infection found that 26.6% of infections had onset during a hospitalization, 58.4% were community-onset infections among individuals with previous health care exposures, and 13.7% were CA infections [9].
CA MRSA strains typically have PFGE patterns (e.g., USA300 or USA400) that differ from HA MRSA strains, are staphylococcal cassette chromosome mec type IV or V, produce Panton-Valentine leukocidin (PVL), and are susceptible to most non-β-lactam antibiotics [6, 10, 11]. The predominant CA MRSA strain in the United States is multilocus sequence type 8 (ST8):USA300:staphylococcal cassette chromosome mec type IVa, whereas the predominant CA MRSA strains in Europe and Oceania are ST80 and ST30, respectively [12, 13]. All predominant CA MRSA strains produce PVL, which is strongly associated with the propensity of these organisms to cause skin and soft-tissue infection, necrotizing fasciitis, and necrotizing pneumonia and with the ability of such strains to spread rapidly [14, 15]. USA300 also carries genes for molecular variants of enterotoxins K and Q and a novel arginine catabolic mobile element that encodes an arginine deiminase pathway and oligopeptide permease system, which may contribute to the epidemiological and clinical virulence of USA300 [14].
CA MRSA strains have been introduced into hospitals by the 15%–23% of patients who require hospitalization for management of their infections [7, 8] and most likely by other colonized patients who require hospital admission for other purposes. In 2003, Saiman et al. [16] reported one of the first recognized outbreaks involving transmission of a CA MRSA strain in a health care facility. Subsequently, a number of nosocomial outbreaks, often occurring in nurseries or maternity wards and less frequently among health care workers, have been reported [17, 18]. In some locations, USA300 has become a relatively common cause of sporadic surgical-site infections and hospital-onset bloodstream infections (BSIs) [19, 20]. The nosocomial outbreaks of CA MRSA infection affecting “low-risk” groups such as newborns and postpartum women are reminiscent of the serious epidemics of methicillin-susceptible S. aureus infections in the 1950s and 1960s that were caused by phage type 80/81 strains [21–23], which have been shown to be ST30- and PVL-positive strains [24]. The extent to which importation of CA MRSA into hospitals has occurred—and whether these strains will cause major changes in the clinical and epidemiological aspects of HA infection—are issues that warrant careful scrutiny.
In this issue of Clinical Infectious Diseases, Popovich et al. [25] describe a study wherein molecular methods (PFGE, PVL, and staphylococcal cassette chromosome mec typing) and antimicrobial susceptibility patterns of a subset of isolates from hospital-onset MRSA BSIs were used to derive a phenotypic rule for classifying MRSA isolates as being either inferred community genotype (CG) or inferred hospital genotype (HG). Validated phenotypic rules of this type may be useful to clinicians and epidemiologists when resources for more-accurate molecular typing of isolates are limited (or nonexistent) in an institution or for retrospective analyses of MRSA trends when isolates are not available for molecular typing [26, 27]. By applying the phenotypic rule to a large number of BSIs over a 7-year period, the authors established that the proportion of MRSA BSIs caused by CG MRSA increased from 24% to 49%. This finding is of particular concern, because it is substantially higher than the 16%–22% of invasive health care-associated infections caused by USA300, as reported in the study by Klevens et al. [9]
Consistent with the known antimicrobial susceptibility patterns of CG MRSA, Popovich et al. [25] also found that the proportion of hospital-onset MRSA BSI isolates that were resistant to clindamycin, gentamicin, and ciprofloxacin decreased significantly during the study period. Interestingly, the overall incidence of hospital-onset MRSA BSI during the study period remained relatively stable, because BSIs due to HG strains decreased in frequency as incidence of CG MRSA increased. Given the fact that PVL-positive CA MRSA strains are widely believed to be more virulent than typical HA MRSA strains, it is of interest that the duration of bacteremia, duration of hospital stay, rate of readmission to the hospital within 3 months, and overall mortality were similar in patients with BSIs due to CG MRSA and in patients with BSIs due to HG MRSA. Although it is somewhat reassuring that BSIs caused by CG strains were not more severe than those caused by HG strains, additional studies are needed to establish the full impact, over time, of CA MRSA strains on HA infections, including skin and soft-tissue infections, surgical-site infections, and lower respiratory tract infections.
As pointed out by Popovich et al. [25], the applicability of the phenotypic rule used in the current study to other health care settings will depend on local antimicrobial susceptibility patterns of MRSA. In the study hospital, isolates susceptible to either clindamycin or ciprofloxacin were classified as being CG (CA MRSA). However, the predominating CA MRSA clone can differ in hospitals within a small geographic area, fluoroquinolone-resistant CA MRSA strains have already emerged in some areas, and susceptibility to clindamycin can change over time [28–30]. As a result, accurate characterization of circulating strains of MRSA will, whenever possible, be best achieved by molecular typing methods. The fact that multilocus sequence typing, PFGE, and staphylococcal cassette chromosome mec and PVL typing of isolates are time-consuming and are not available in a majority of clinical laboratories highlights the need for rapid and more-simplified methods of identifying clones of CA MRSA [31]. However, as CA MRSA strains become more common in hospitals, as noted in the study by Popovich et al. [25] and elsewhere [9], the distinctions between CA MRSA and traditional HA MRSA clones may become less relevant.
Continued importation and transmission of PVL-positive CA MRSA strains in health care facilities will likely make control of HA MRSA infection increasingly difficult [17]. Accordingly, there is an urgent need to improve adherence to infection-control measures designed to minimize the spread of MRSA infection. High priority should be given to improving hand hygiene among health care workers, which is widely believed to be one of the most important strategies for controlling HA infection, including MRSA infection [32, 33]. Personnel should be made aware that patients infected with MRSA are usually colonized at multiple body sites (including normal-appearing areas of skin) and that they often contaminate items in their immediate vicinity [34–36]. For these reasons, health care workers must improve their adherence to recommended barrier precautions (gloves and gowns), which have been confirmed in a recent quasiexperimental study to be effective in reducing transmission of MRSA infection [37, 38]. Because contaminated environmental surfaces in the rooms of affected patients may contribute to transmission of MRSA infection [39–41], greater emphasis on effective daily and terminal cleaning and disinfection of the environmental surfaces in rooms occupied by MRSA-infected patients is also warranted [37, 42]. Performing surveillance cultures on patients at risk of MRSA colonization is another important element of MRSA-control programs in health care facilities [37, 43]. Screening the anterior nares of individuals at the time of hospital admission will detect colonization in most patients with HA MRSA strains. However, some individuals with CA MRSA skin and soft-tissue infections have not been colonized in the nares at the time that investigations were conducted [44, 45]. In one study, HIV infection and a clinical diagnosis of a skin or soft-tissue infection, in addition to exposure to a health care system, were identified as risk factors for MRSA colonization at the time of hospital admission [46]. Screening patients' anterior nares and any skin lesions or wounds (or groin and/or axillary areas, if no wound is present) should identify most colonized patients. Tests with new chromogenic MRSA-selective media or PCR methods identify colonized individuals more rapidly than do culture of specimens inoculated onto traditional agar media [47, 48]. Incorporation of all of the measures listed above into a hospital MRSA-control program is likely to provide the best protection against the increasing threat of MRSA infection.
Acknowledgments
Potential conflicts of interest. J.M.B. has received honoraria from BD GeneOhm.
- Received December 3, 2007.
- Accepted December 4, 2007.
- © 2008 by the Infectious Diseases Society of America
References
This Article
- ExtractFree
- » Full Text (HTML)Free
- Full Text (PDF)Free
Classifications
-
- Articles and Commentaries
Services
Citing Articles
Google Scholar
PubMed
Related Content
Share
- Email this article
-
CiteULike
Connotea
Delicious
Facebook
Google+
Mendeley
Twitter
Published on behalf of
IDSA Members: For your free access to this journal, log in via the IDSA members area.
The Journal
Impact Factor: 8.186
5-Yr impact factor: 7.898
For The Media
For Authors
- Services for authors
- Self-archiving Policy
- IDSA Conflict of Interest policy
- Instructions for Authors
- ICMJE Form
- Submit Now!
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy
Corporate Services
Alerting Services
Related IDSA Publications
