S316 17th ECCMID / 25th ICC, Posters highly sensitive detecting 3.3ng of F1 CA and was the most convenient and economic test. In contrast to immunofluorescence microscopy, ELISA can be used as a screening tool and for quantification of F1 CA. Flow cytometric analysis is sophisticated and too expensive for laboratories in most areas where plague is endemic. Commercially available products will become more attractive when certified for diagnostic purposes and available at lower costs. P1160 Development of a rapid fluorescence assay in 96 well plate as- say for multiple antibiotic resistant Salmonella typhimurium N. Coldham, L. Randall, L. Piddock, M. Woodward (Surrey, UK) Objectives: Multiple antibiotic resistant (MAR) mutants of Salmonella enterica and E. coli are characterised by MIC increases of 2-8 fold to unrelated antibiotics including tetracycline, chloramphenicol, some penicillins and quinolones. They have been associated with clinical disease in their own right and provide a platform for progression to high level antibiotic resistance. Previous studies have indicated that the AcrA/B-TolC efflux pump system and certain porins cooperate to reduce the uptake of antibiotics and are important effectors of MAR. Many substrates of AcrA/B-TolC are divalent cations and include fluoroquinolone antibiotics and ethidium bromide. In the present study, intracellular concentration of the fluorescent probe bis benzimide (H33342) was evaluated as a rapid single test for MAR Salmonella Typhimurium. Methods: S. Typhimurium (SL1344), isogenic MAR mutants (n = 4) and isogenic mutants with defined deletions in acrB, tolC, ompC and ompF were cultured overnight in LB broth and diluted with phosphate buffered saline to an optical density of 0.1 (600 nm). The strains were incubated in 96 plate wells with bis benzimide (2.5 mM) and the fluorescence recorded for 30 minutes using a Fluostar plate reader at excitation and emission wavelengths of 350 and 460 nm respectively. Sensitivity of these strains to the efflux pump inhibitors Phe-Arg-b-naphthylamide (166 ug/mL) and carbonyl cyanide m-chlorophenylhydrazone (2 ug/mL) was also evaluated. The cyclohexane tolerance of all strains was also determined. Results: Only the MAR mutants were tolerant to cyclohexane. Significantly (P < 0.0001) reduced uptake of bis benzimide was observed by the MAR mutants and strains with defined gene deletions in ompC and ompF compared to their parent (SL1344). The uptake of bis benzimide was unchanged in the acrB deletion mutant presumably due to increased compensatory expression of AcrE and AcrF efflux pump proteins. Increased uptake of bis benzimide was observed in the tolC gene deletion mutant and heat inactivated controls. Uptake by all strains was inhibited by the efflux pump inhibitors. Reduced sensitivity of the MAR mutants to CCCP further amplified the difference in bis benzimide uptake compared to the parent strain. Conclusion: A rapid test in 96 well plate format has been developed and partially validated for the detection of MAR mutants. Differential sensitivity to efflux pump inhibitors enables further concurrent test amplification and phenotypic analysis. P1161 Comparison of selective media for the detection and enumeration of lactobacilli and bifidobacteria from human faecal samples V. Vankerckhoven, C. Lammens, O. Chonan, K. Oishi, H. Goossens (Wilrijk, Ghent, BE) Objectives: Methods for the detection and enumeration of faecal bacteria should be properly validated. The aim of this study was to compare selective media for the enumeration of lactobacilli and bifidobacteria from human faecal samples. Methods: Serial dilutions of 15 human faecal samples were plated, using the spiral enter apparatus Eddy Jet, on two commonly used selective media, LAMVAB and LBS for the detection and enumeration of lactobacilli, and transgalactosyloligosaccharide agar (TOS) and modified TPY agar (MTPY) for bifidobacteria. Colonies were counted based on colony morphology and Gram stain after a 72 h anaerobic incubation at 37ºC. The number of lactobacilli and bifidobacteria per gram wet weight of faeces was estimated from the number of colonies. Results: Total colony counts ranged from <2.0 to 5.8 log10 cfu/g faeces for LAMVAB and from <2.0 to 5.5 log10 cfu/g faeces for LBS. Mean bacterial counts for lactobacilli were 5.0 log10 cfu/g faeces on LAMVAB compared to 4.9 log10 cfu/g faeces on LBS (P > .05). Using LAMVAB, lactobacilli were found in 11 of 15 (73%) faecal samples compared to 9 of 15 (60%) using LBS (P > .05). For the bifidobacteria, total colony counts ranged from 7.1 to 9.6 log10 cfu/g faeces for TOS and from 7.0 to 9.4 log10 cfu/g faeces for MTPY. Mean bacterial counts for bifidobacteria were 8.3 log10 cfu/g faeces on both TOS and MTPY (P > .05). Bifidobacteria were found in all (100%) faecal samples using both media (P > .05). Conclusions: It can be concluded that both lactobacilli and bifidobacteria could reliably be detected on each of the two selective media. For lactobacilli, LAMVAB and LBS showed comparable results and could be used to quantify lactobacilli in human faeces. The same holds for the media TOS and MTPY, which showed comparable results and could be used for the quantification of bifidobacteria from human faecal samples. P1162 Retrospective analysis of Staphylococcus lugdunensis isolates K. Chatzigeorgiou, N. Siafakas, A. Tarpatzi, S. Chaniotaki, S. Kouvardas, L. Zerva (Athens, GR) Objectives: Staphylococcus lugdunensis is an emerging pathogen of a yet undefined clinical importance. The aim of the present study was the retrospective analysis of all S. lugdunensis isolates recovered in a Microbiology Laboratory and the assessment of their clinical significance. Methods: Twenty-five Gram (+) cocci initially identified by Phoenix (Becton Dickinson) as S. lugdunensis during a 31 month period were obtained from 12 patients. Thus, 12 strains (1 strain per patient) were thawed and retested by Phoenix. API Staph system (bioM´ erieux), 16S rRNA gene sequencing, ornithine decarboxylase (ODC), PYR (Remel), slide (BD) and tube (Remel) coagulase tests were additionally performed. Susceptibility tests included disk diffusion (penicillin, oxacillin, cefoxitin, vancomycin, teicoplanin, tetracycline, gentamicin, erythromycin, clindamycin, ciprofloxacin, norfloxacin, rifampin and nitrofurantoin), E-test (penicillin and oxacillin), mecA gene (PCR) and PBP2a protein (slide latex, bioM´ erieux) detection and b-lactamase production (Oxoid). All relevant clinical information was retrieved from the medical records. Results: By 16S rRNA sequencing, 10 of the 12 strains were identified as S. lugdunensis; the rest were not further considered. API Staph testing produced identical results with sequencing, while Phoenix misidentified 1 as S. capitis ssp. ureolyticus. All were ODC and PYR (+), tube coagulase (-) and half bound coagulase (+). All were mecA and PBP2a (-) and one produced b-lactamase. They all exhibited a multi- susceptible phenotype except for 1 gentamicin- and 1 quinolone-resistant strain. Clinical presentation included 1 case of endocarditis and 2 bacteraemias with excellent clinical outcome. The other isolates were obtained from pus (6 cases) or urine (1 case) specimens and their clinical significance was doubtful. Finally, analysis of laboratory records demonstrated that S. lugdunensis represented 1.3% of all coagulase negative isolates. Conclusion: S. lugdunensis can be misidentified during routine laboratory practice and supplementary biochemical asays, especially ODC and PYR, are crucial for its recognition. It is rather a multi- susceptible species and may cause severe infections, albeit with good prognosis, or simply represent a contaminant.