Background Highly specific and sensitive discrimination between closely related pathogenic and commensal Neisseria spp is crucial because these species frequently colonise the same anatomical sites. Herein, two commercially available Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (MALDI-TOF-MS) platforms and one independent software and database (Andromas) was compared to conventional phenotypic and genetic tests routinely used for identification of Neisseria spp.

Methods The performance of each platform, analysing 129 pathogenic isolates (Neisseria gonorrhoeae and Neisseria meningitidis) and 69 commensal Neisseria isolates (15 different species), was determined by deposition of one single culture colony to the MALDI plate and analysed in Microflex (Bruker, Germany) and VITEK MS (bioMérieux, France) according to the manufacturer’s instructions. Subsequently, the acquired data from Microflex was submitted for analysis in the Andromas software and database (Andromas, France), which uses a different algorithm for species identification. Unfortunately, VITEK MS data is not compatible with Andromas. Phenotypic and genetic (16S rRNA gene sequencing) methods were used for final discrepancy analysis (still pending).

Results Microflex correctly identified all (100%) N. gonorrhoeae and N. meningitidis, however, four commensal isolates were indicated as possible N. meningitidis. Three of these four isolates were N. kochii. The VITEK MS misidentified 1 N. gonorrhoeae, 1 N. meningitidis and 2 commensal isolates (both N. kochii) were indicated as possible N. meningitidis. Finally, Microflex data analysed in the Andromas software and database correctly identified all (100%) pathogenic and commensal strains.

Conclusion This study shows that both Microflex and VITEK MS discriminate pathogenic Neisseria species from commensal Neisseria species with a high, but not ideal, specificity. Furthermore, an optimal MALDI-TOF-MS platform should be compatible with secondary softwares and databases for confirmation. Importantly, the Microflex results analysed in the secondary software and database Andromas correctly identified all (100%) pathogenic and commensal strains.