Comparison of two matrix-assisted laser desorption ionisation-time of flight mass spectrometry methods for the identification of clinically relevant anaerobic bacteria.

Two commercially available MALDI-TOF MS systems, Bruker MS and Shimadzu MS, were compared for the identification of clinically relevant anaerobic bacteria. A selection of 79 clinical isolates, representing 19 different genera, were tested and compared with identification obtained by 16S rRNA gene sequencing. Correct genus identification was achieved for 71% of isolates by Shimadzu MS and for 61% by Bruker MS. Correct identification at the species level occurred in 61% and 51%, respectively. Shimadzu showed markedly better results for identification of Gram-positive anaerobic cocci. In contrast, the Bruker system performed better than Shimadzu for the Bacteroides fragilis group. When strains not present in the database were excluded from the analyses for each database, both systems performed equally well, with 76.7% and 75.0% correct genus identification for Shimadzu and Bruker, respectively. Similarly, when the most recently updated Bruker database was applied, no difference was observed. We conclude that the composition and quality of the database is crucial for a correct identification. The databases currently available for both systems need to be optimized before MS can be implemented for routine identification of anaerobic bacteria.