Methicillin-resistant S. aureus (MRSA) frequently causes disease outbreaks and has become endemic in many regions, adding to the morbidity, mortality, and cost of care associated with hospital-acquired infections. Enhanced surveillance and infection control measures have been adopted by healthcare institutions (HCIs) to address this unresolved problem (5). In particular, reporting of bloodstream infections (BSI) by MRSA is often mandatory and reduction of BSI rates is a performance target (5,12,21).In the Centre region of France, an extensive, prospective, longitudinal, region-wide survey of BSI has been under way since 2000. Data are collected for 3 months of each year in a large number of HCIs to establish a comprehensive picture of the epidemiology of severe hospital-acquired infections. MRSA BSI and methicillin-sensitive S. aureus (MSSA) BSI are extensively studied within this framework. All of the S. aureus strains isolated during successive study periods are sent to our central laboratory for susceptibility testing, molecular typing, and analysis of virulence genes with the aim of determining the spread and diversity of S. aureus strains in the region. The results obtained during the first 4 years of surveillance (2000 to 2003) of MRSA BSI have been reported previously (27).Here we report the data from 2004 to 2006. We looked for any major changes in the epidemiology of antibiotic resistance and of virulence genes in strains of S. aureus responsible for BSI. We identify a need to focus efforts on preventing both MRSA and MSSA BSI infections and raise the issue of whether the use of fluoroquinolones (FQs) has contributed to the acquisition of resistance and virulence genes by S. aureus strains.
MATERIALS AND METHODSBSI epidemiological survey method. A BSI surveillance program in the Centre region of France (2.5 million inhabitants) and a microbiological study of S. aureus strains isolated from BSI cases have been conducted since 2000. Thirtytwo HCIs, comprising 6,027 short-stay beds, participated in this annual 3-month survey of all cases of BSI. Here, we report results for the years 2004 to 2006. The survey covered 2,007,681 patient days (PD). The methods, study design, and data for the years 2000 to 2003 have been reported elsewhere (27). Briefly, the variables studied included patient age and sex, portal of entry, community-or hospital-acquired BSI, occurrence of death within 7 days of BSI diagnosis, and duration of hospital stay. Data were analyzed with Epi Info v.6 software. Data were analyzed with a 2 test with five degrees of freedom. The incidences of community-acquired and nosocomial BSI were determined with respect to the number of PD.Microbiological methods. (i) Bacteriology. Three hundred fifty-eight BSIassociated S. aureus strains were collected during the three survey periods (2004, 2005, and 2006). The strains were sent to the reference laboratory of the Relais d'Hygiène du Centre. The isolates were identified as S. aureus according to previously described procedures (27).(ii) Antimicrobial ...