[Comparison of two reverse hybridization methods for mycobacterial identification in clinical practice]

To evaluate the performance of two commercial methods for identification of Mycobacterium species: InnoLiPA Mycobacteria first version (Innogenetics) versus Genotype MTBC and Genotype Mycobacterium (HAIN) on, respectively, 2123 and 2164 distinct isolates.

Both techniques are based on the reverse hybridization of PCR products to their complementary probes immobilized on membrane strips. The InnoLiPA assay targets the 16S-23S rRNA spacer region. The HAIN test is composed of two kits: Genotype MTBC, for identification of tuberculosis complex mycobacteria, is based on gyrB DNA sequence polymorphism. Genotype Mycobacterium kit targets the 23S rDNA for identification of mycobacteria other than tuberculosis (MOTT) and tuberculosis complex mycobacteria. Both assays identify complex tuberculosis mycobacteria and respectively, eight and 12 species of MOTT. Moreover, the Genotype MTBC allows species differentiation within the M. tuberculosis complex.

Eighty-eight percent and 95% of mycobacteria were identified by InnoLiPA and HAIN, respectively. Hybridization remained negative for 11% of isolates with InnoLiPA and 4% with HAIN. An identification of MOTT was obtained by conventional identification in all cases after the use of InnoLiPA. MOTT and one M. tuberculosis was obtained after HAIN procedure. Unidentified species were complementary to a specific probe in 5% of the cases with InnoLiPA and 17% with HAIN.

HAIN identifies more mycobacteria species than does InnoLiPA and allows identification in the M. tuberculosis complex. However, failure in identification occurs only with MOTT with InnoLiPA when one M. tuberculosis was found among mycobacteria non identified with HAIN.