S236 17th ECCMID / 25th ICC, Posters PCR assay targets the 3 ′ integration site of the staphylococcal cassette chromosome mec (SCCmec) at the orfX locus. In the present study we improved our real time PCR to detect in addition to SCCmec types I–IV the type V cassette. We demonstrate here that 226 MRSA, 111 methicillin-sensible Staphylococcus aureus (MSSA), 50 methicillin-resistant coagulase-negative Staphylococci and 12 methicillin-sensible coagulase-negative Staphylococci were correctly identified. However, 3 MSSA were false positive and the reason for this observation remains currently unknown. Furthermore, we evaluated our method with nasal and wound swabs from patients. In our ongoing clinical study we have analyzed 519 swabs by culture and PCR. 489 swabs were negative in both experimental procedures. All together 30 swabs contained MRSA. 15 swabs had positive results in PCR and culture method, 8 swabs were only in PCR positive and 7 only by culture. The swabs, which were only positive in PCR, were confirmed to be MRSA-positive in an independent PCR-assay detecting mecA and a Staphylococcus aureus-specific gene. To rule out the possibility that our SCCmec-PCR was unable to detect the 7 MRSA- strains that were only positive in culture, we repeated the PCR-assay in 5 out the 7 cases (in 2 cases the strains were unfortunately not available) from colony material. In all 5 cases the SCCmec-PCR was positive. The earlier analysis (3 hours versus at least 5 days) brings an overall cost reduction in the clinic because hygiene measures can be employed quicker and the spread of MRSA thus prevented. Isolation measures of patients with former MRSA-carriage can be finished sooner, if the SCCmec-PCR assay is negative. In conclusion the SCCmec-PCR is a reliable and time saving detection method for MRSA in the clinic and our studies show that this PCR assay equals the accuracy of culture. P908 The ability of a rapid PCR-based method to identify surgical patients colonised with methicillin-resistant Staphylococcus aureus K.J. Hardy, S. Gossain, C. McMurray, S. Shabir, I. Alobwede, A. Williams, N. Athamneh, P.M. Hawkey (Birmingham, UK) Objectives: To determine the ability of the rapid IDI MRSA PCR based assay to identify surgical patients colonised with MRSA compared to conventional culture methods. Methods: During a 9 month period all patients admitted to 4 surgical wards had nasal swabs taken for detection of MRSA on admission and every 4 days subsequently. Each swab was plated directly onto MRSA ID agar and then tested using the IDI MRSA assay. Additionally all swabs were placed in brain heart infusion broth incubated overnight and subcultured onto MRSA ID and Baird Parker agar for detection of MRSA and MSSA respectively. Results: A total of 7545 nasal samples were examined from 3769 patients. Seven hundred and fifty six samples were positive by the IDI MRSA assay, of which MRSA was isolated on direct culture from 374 (49.5%). Of the 382 samples that were negative on direct culture, MRSA was isolated on broth enrichment from 138 samples, and MSSA from 42 samples. Of the 244 samples that were IDI MRSA assay positive but did not have MRSA isolated, 77 (31.6%) came from patients that had previous or subsequent samples from which MRSA was isolated. When individual patients are examined a total of 465 of the 3769 patients (12.3%) sampled had at least one sample that was positive by the IDI MRSA assay, of which 254 (54.6%) were positive by direct culture. An additional 76 patients had MRSA isolated from broth enrichment, but a total of 135 patients (29%) were only ever found to be positive by the IDI MRSA assay and did not have MRSA isolated using culture. Overall the IDI MRSA assay has a sensitivity of 97%, a specificity of 96.1%, NPV of 99.7% and a PPV of 71%. Conclusion: The IDI MRSA assay provides a rapid and sensitive way of screening surgical patients for MRSA, with a very high negative predictive value. The increased number of patients identified as colonised with MRSA using the IDI MRSA assay compared to conventional culture, may be due to the increased sensitivity of PCR compared to culture. Rapid and sensitive identification of surgical patients colonised with MRSA aids in the prompt decolonisation and isolation of patients. P909 Genome variability of methicillin-resistant Staphylococcus aureus as revealed by comparative genome hybridisation with a multi-strain microarray G. Kuhn, T. Koessler, P. Francois, A. Huyghe, J. Schrenzel, P. Francioli, D.S. Blanc (Lausanne, Geneva, CH) Objectives: Using comparative genome hybridisation (CGH), we firstly aimed to analyse core and accessory genome compartments in a diverse collection of Methicillin resistant S. aureus (MRSA). Secondly, we aimed to know whether genetic elements have specifically been gained or lost during evolution of S. aureus Clonal Complexes (CC’s). Methods: Our collection comprised 21 MRSA isolates of 5 different CC’s. It included 1 isolate of CC1, 6 isolates of CC5, 7 isolates of CC8, 1 isolate of CC22, 7 isolates of CC45 and 2 singletons (ST 80 and 15). CGH was performed on a multi-strain oligonucleotide microarray. It comprised probes for S. aureus strains N315, Mu50, COL, and MW2 covering 99% of all genomes with 2535, 2663, 2633, and 2625 genes respectively. Based on genome sequences of all 4 strains, 2019 genes are core. Oligonucleotide probes included common probes targeted at conserved gene regions (based on alignment of orthologs of all 4 strains) and specific probes targeted at variable gene regions. Common probes were designed to avoid absence of signals in CGH due to sequence variation. For analysis of S. aureus CC evolution we subtracted isolates of CC5, CC8, or CC45 from all other isolates. Results and Discussion: CGH of DNA from 21 strains to common probes showed that the core genome comprised only 1851 genes. Interestingly, we noted that a considerable number of genes, which were located on mobile genetic elements, were present in all strains. In contrast to this, other genes, which were clearly not associated with mobile genetic elements, were variably lost. A phylogenetic analysis based on presence and absence of accessory genes was performed and showed a weak congruence with CC groups and this was especially true for isolates of CC8. In line with this observation, we did not find genes which were specifically associated with either CC5, CC8, or CC45. This was in contradiction to previous studies using CGH. Conclusions: By performing CGH analysis on 21 genetically diverse strains using a microarray which comprised a set of common probes we were able to clearly differentiate core and accessory genes. Our results show that accessory genes are not useful for phylogenetic analyses. P910 Rule-out Staphylococcus aureus, including MRSA, directly from primary swab cultures using S. aureus PNA FISH A. Voss, M. Wulf, M. Sørum, R. Skov, K.G. Madsen (Nijmegen, NL; Copenhagen, Vedbæk, DK) Objectives: In July 2006, we performed a study among veterinarians attending the International Pig Veterinary Society Congress (IPVS) in Copenhagen. This study was set-up to screen the attendees for MRSA carriership and at the same time to validate S. aureus PNA FISH (AdvanDx Inc.) as a rapid rule-out test from primary swab cultures. S. aureus PNA FISH is a rapid molecular diagnostic test with a 2.5 hours turn-around time once culture swab culture reach the laboratory. Methods: 264 swabs from the anterior nares and throat (combined culture) were collected from participants at the IPVS Congress. The swabs were transported to Statens Serum Institut for over-night enrichment in salt-enriched semi-selective nutrient broth. One ml of this primary swab culture was frozen and shipped to Nijmegen, The Netherlands for analysis by S. aureus PNA FISH. Subsequently, the broth was subcultured on blood agar and MRSA-ID agar plates (bioM´ erieux). Staphylococcus aureus was identified by colony morphology and tube coagulase test. Oxacillin-resistance was determined by cefoxitine-disc- diffusion according to CLSI criteria and confirmed by in-house mecA PCR.