A Retrospective Review of MRSA Susceptibilities from Long Term Care Facility Patients by REP-PCR Typing

Background: Methicillin resistant Staphylococcus aureus (MRSA) is a frequent cause of significant infections in the community, acute care, and long-term care facility (LTCF) settings.  As part of a surveillance project funded by the Center for Disease Control and  Prevention Epicenter Program, MRSA isolates from an academic medical center and 7 outreach hospitals were collected between 1/2007 and 6/2010.  Cases of patients with recent LTCF stays were identified.    

Objective: To characterize the molecular genotype, patient characteristics, and clinical outcomes of MRSA strains causing infection in recent LTCF patients included in this surveillance cohort.  

Methods: Patients were included if they had been in a LTCF within the preceding 4 weeks before the culture was obtained.  Molecular genotyping by repetitive element Polymerase Chain Reaction (rep-PCR;DiversiLab,TM Biomerieux, NC) was conducted on MRSA isolates strains collected from recent LTCF patients.  Pulse field gel electrophoresis (PFGE) types were derived from DiversiLab MRSA library and performed on selected representative repPCR isolates.  We compared the strain types of MRSA found in each LTCF 5 digit ZIP code areas with the clinical outcome, demographics, and antimicrobial resistance patterns from respective patients.  Data was collected through medical record reviews. Clustering analysis through geocoding (ArcGIS 9.3) and K-means clustering are used to measure the significance of the findings. 

Results: Among 163 isolates from recent LTCF residents, 63% (95/163) were PFGE type USA 100, 18% (30/163) USA 300, 13% (22/163) Brazilian type, and 5% (8/163) others.  Of the 35 documented rep-PCR types, 8 predominated. There was a trend toward increased cure rates with the traditionally community acquired USA 300 versus USA100 and Brazilian types (43% vs 28% and 35%, respectively), but this could be partially explained by an average age that was 10 years greater in the USA 100 patients.  Site of infection varied among MRSA strain types; Brazilian strains were twice as likely (28% vs 13%) to cause pneumonia as compared to the USA 100 strain, but not associated with documented cases of skin and soft tissue infection.  Mapping (geocoding) cases by the preadmission LTCF zip code suggests clustering of strain types.  Susceptibility data from these rep-PCR defined clonal types can provide clinical guidance on initial empiric treatment based on location prior to patient transfer.

Conclusions: Most recent LTCF patients were infected with MRSA strains commonly associated with healthcare exposure (USA 100, Brazilian). A smaller proportion had USA 300 strains indicating the introduction of this strain into the LTCF environment.  By mapping strain locations regional clonal dissemination is suggested.  This information is clinically valuable in predicting treatment course and site of infection on presentation.