Increase in the Reservoir of Community-Acquired MRSA, With Implications for Hospital-Acquired Infection
Abstract & Commentary
Synopsis: Over a 7-year period, there was a dramatic increase in the isolation of community-acquired MRSA in the San Francisco area. Molecular typing showed movement of community-acquired strains into hospitals.
Source: Carleton HA, et al. Community-Adapted Methicillin-Resistant Staphylococcus aureus (MRSA): Population Dynamics of an Expanding Community Reservoir of MRSA. J Infect Dis. 2004;190:1730-1738.
Community-acquired MRSA (CO-MRSA) have been isolated with increasing frequency over the past decade. One possible explanation is that many strains of CO-MRSA are in fact nosocomial MRSA (NO-MRSA) that have entered into the community. However, a number of strains of CO-MRSA have genetic and phenotypic characteristics that differentiate them from NO-MRSA. True CO-MRSA typically carry the type IV staphylococcal chromosomal cassette (SCC) mec element that confers beta-lactam resistance. NO-MRSA usually carry the type II SCCmec element which encodes beta-lactam resistance, but typically carries resistance to other unrelated antimicrobials as well. Thus, CO-MRSA tend to be susceptible to multiple non-beta-lactam agents, whereas NO-MRSA are multi-resistant.
Carleton and colleagues studied the molecular epidemiology of MRSA in the hospital and community over a 7-year period. They collected isolates from a university hospital in San Francisco, as well as affiliated long-term care, correctional, and outpatient facilities. They performed molecular typing, including SCCmec typing, pulsed field gel electrophoresis (PFGE), multilocus restriction-fragment typing (MLRFT), and multilocus sequence typing (MLST) on a random sample of nearly 500 isolates. Between the years 1998 and 2002, they found a 4-fold increase in the incidence of MRSA. The incidence of NO-MRSA (defined as isolation of MRSA > hours after hospital admission) during the period was relatively stable. There was a large and statistically significant increase in the incidence of CO-MRSA, however. Concomitantly, there was a parallel increase in the rate of isolation of type IV SCCmec-bearing strains, with most of these belonging to 4 predominant genotypes. These 4 genotypes were predominantly associated with community-acquired disease (76.9%), rather than disease acquired in the hospital (19.4%) or long-term care facility (3.7%). One of these strains accounted for 30% of CO-MRSA; after it became established in the community it was isolated from cases of NO-MRSA. By the end of the survey period, it accounted for 14% of NO-MRSA.
Carleton et al noted that some of the isolates from NO-MRSA cases belonged to genotypes associated with community-acquired disease, but bore type II SCCmec, and were resistant to multiple non beta-lactam antimicrobials. Type II SCCmec isolates were largely confined to the hospital, indicating limited ability to spread in the community.
Comment by Robert Muder, MD
CO-MRSA appears to be increasing in frequency in numerous geographic areas. A number of epidemiologists have identified prior admission to a hospital, or other contact with the healthcare system as a risk factor for CO-MRSA. Such infections are not truly community-acquired, but rather community onset. Carelton et al took advantage of a major genetic difference between nosocomial and community-acquired strains. Strains of nosocomial origin in the United States usually contain type II SCCmec, which typically encodes resistance to multiple antimicrobials in addition to beta-lactam resistance. Strains of community origin typically contain SCCmec IV which does not. One explanation is that the genetic cost of carrying these additional determinants makes type II SCCmec poorly adapted to the community setting, in which multiple antimicrobials are less likely to be encountered. Thus, they tend to stay within healthcare facilities, where heavy antibiotic pressure confers on them a survival advantage.
Carelton et al were able to show that the increase in MRSA noted in their community was largely due to an increase in community-acquired strains. Further, the community strains were able to enter the hospital and displace strains of nosocomial origin. Some of these community strains appeared to adapt to the hospital environment by acquiring the SCCmec II genotype, conferring additional resistance determinants.
These findings have important implications for both public health and for control of nosocomial infection. CO-MRSA is clearly an emerging public health problem that is occurring independently of what is happening in hospitals. The expected spread of NO-MRSA into the community does not seem to have materialized. On the contrary, CO-MRSA strains appear to be moving into the hospital and acquiring additional resistance determinants. This is likely to complicate attempts at control of NO-MRSA to a considerable degree.
Robert Muder, MD, Hospital Epidemiologist Pittsburgh VA Medical Center Pittsburgh Section Editor, Hospital Epidemiology, is Contributing Editor for Infectious Disease Alert.