Species identification with the BD Phoenix^® System at low and high inoculum in a routine laboratory

Abstract number: 1733_1217

Geppert F., Kempter N.

Objectives: Most automated systems require inoculation with McFarland 0.5 (1.5×10⁸ CFU/mL). In the routine laboratory sometimes only few colonies resulting in lower applicable inocula are available. Becton & Dickinson has thus developed a new software allowing a lower density of McFarland 0.25 (7.5×10⁷ CFU/mL). We compared the results obtained with these two inoculation modes under routine conditions.

Methods: 181 Gram-positive and Gram-negative isolates including fastidious and difficult to identify strains from clinical material were analysed. Reference diagnoses were made by conventional methods using the API- and ID 32 system, tube coagulase (RAPIDEC^®) and molecular typing (GenoType MRSA^®, Hain Lifescience). Using the Phoenix^®, all strains were tested in parallel at the standard inoculum of McFarland 0.5 (P 0.5) and 0.25 (P 0.25), respectively, after suspension in 4.5 ml of identification broth. The P 0.25 panels were marked for automatic identification by the system and the results were evaluated in separate data bases.

Results: 80% of the ID results were available within 4 hours. Accurate identification (ID) at species level was achieved in 94.5% at P 0.5 and in 93.4% at P 0.25, respectively. Details of discordant results (n = 18) are presented in the table.

Conclusion: The new software accurately identified strains at a lower inoculum, interestingly in some cases even better than with McFarland 0.5. Since susceptibility testing can be done simultaneously and there is no need for subculturing in case of low biomass, a report is completed 18–24 hours earlier.