Abstract
OBJECTIVE: Mycobacterium tuberculosis isolates that fail to hybridize to at least one rpoB wild-type or any mutation probe on the Genotype MTBDRplus strip are assumed to be rifampicin-resistant. However, the precise mutation(s) are unknown. We sought to identify the mutations in isolates with such hybridization patterns and determine if the mutations are associated with resistance to rifampicin. METHODS: In this study, 275 M. tuberculosis isolates were screened with the Genotype MTBDRplus assay to identify isolates with the hybridization pattern. These isolates were sequenced and their minimum inhibitory concentrations (MIC) determined using the Bactec MGIT 960 system. RESULTS: Among the 275 isolates tested, 15 (6%) isolates with the hybridization pattern were identified. Sequencing showed that failure to hybridize to rpoB wild-type probes resulted from the presence of 'disputed' rifampicin mutations, which are mutations not always associated with a rifampicin-resistant phenotype. All, except 3/15, isolates had a rifampicin-resistant phenotype (MIC > 1 μg/mL). One of the three isolates with a rifampicin-susceptible phenotype had the same mutation at position 526 (His526Leu) as another isolate that had a rifampicin-resistant phenotype. CONCLUSION: The recommendation of the Genotype MTBDRplus assay to assume rifampicin resistance based solely on failure to hybridize to rpoB wild-type probe allows the identification of important RIF-resistant isolates. About 20% (3/15) of such isolates could be missed by relying only on the standard MGIT 960 DST assay for drug susceptibility testing.