Retrospective and prospective evaluation of the FluoroType?-Mycobacteria VER 1.0 assay for the identification of mycobacteria from cultures in a French center

Purpose Rapid, reliable identification of mycobacteria from positive cultures is essential for patient management, particularly for the differential diagnosis of Mycobacterium tuberculosis complex (MTBC) and nontuberculous mycobacteria (NTM) species. The aim of the present study was to evaluate a new "In-Vitro-Diagnostic"-certified PCR kit, FluoroType (R)-Mycobacteria VER 1.0 (Hain Lifescience GmbH) for NTM and MTBC identification from cultures. Methods Mycobacteria identification isolated from positive cultures during routine practice at the Lyon university hospital mycobacteria laboratory obtained by hsp65 amplification/sequencing were compared retrospectively and prospectively to those obtained by and the FluoroType (R)-Mycobacteria VER 1.0 kit. Results The overall agreement between hsp65 amplification/sequencing and the FluoroType (R)-Mycobacteria VER 1.0 kit was 88.4% (84/95); 91.2% (52/57) for the retrospective period and 84.2% (32/38) for the prospective period. There were 9 (9.5%) minor discrepancies (species in the FluoroType (R)-Mycobacteria VER 1.0 database and identified at genus level): 4 during the retrospective period, 5 during the prospective period; and 2 (2.1%) major discrepancies (species in the FluoroType (R)-Mycobacteria VER 1.0 database and identified incorrectly to species level): 1 during the retrospective period (M. kumamotonense identified as M. abscessus subsp massiliense by the kit) and 1 during the prospective period (M. chimaera identified as M. smegmatis by the kit). Including concordant results at genus level and minor discrepancies, 17.9% (17/95) of strains were identified as Mycobacterium sp. by the FluoroType (R)-Mycobacteria-VER 1.0 kit. Conclusion The good performance of the FluoroType (R)-Mycobacteria-VER 1.0 kit with few major discrepancies could enable its use for first-line identification of positive mycobacteria cultures. However, an alternative identification method at least for reference laboratories is needed owing to the non-negligible proportion of NTM strains were identified at genus level.