Epidemiology of methicillin-resistant Staphylococcus aureus (MRSA) bloodstream infection: secular trends over 19 years at a university hospital.

Methicillin-resistant Staphylococcus aureus (MRSA) bloodstream infection (BSI) is a cause of concern in health systems all over the world, due to the high incidence rates and the associated undesirable outcomes. In our tertiary 900-bed university hospital, all episodes of MRSA-BSI have been prospectively followed up since the identification of the first episode in 1990. We conducted the current study to report changes in the epidemiology of MRSA-BSI over the 19-year period between 1990 and 2008, comparing 4 periods (1990-1994, 1995-1999, 2000-2004, and 2005-2008). Overall, 524 patients developed MRSA-BSI. Cumulative incidence was 10.0 episodes/100,000 patient days (range, 1.3-17.4). Although no trend in the incidence rate was observed between the 4 consecutive periods, significant upward trends in patient age and comorbidities, health care acquisition, and non-intravascular catheter source were all identified (p < 0.05). While the Iberian clone (ST247/SCCmecI) was dominant during the first and second periods, almost all the strains isolated in the subsequent periods belonged to Clonal Complex 5 (ST125/SCCmecIV and ST228/SCCmecI). A significant downward trend in vancomycin geometric minimum inhibitory concentration (MIC) was also observed from 2.04 mg/L to 0.88 mg/L, coinciding with the clonal replacement and the reduction in the hospital vancomycin prescription. Therefore, no MRSA vancomycin MIC creep was observed since higher MICs were associated with strains belonging to the Iberian clone. Glycopeptides were the most frequently used antibiotics for MRSA-BSI during all 4 periods. No differences in MRSA-BSI outcomes were found, and the mortality rate at 30 days was close to 29% in each of the 4 periods. In conclusion, we identified significant changes in demographic and clinical characteristics and in the molecular epidemiology of MRSA-BSI during the study period, but found no significant trends in cumulative incidence or in overall mortality rate.