Clinical implications of discrepant results between genotypic MTBDR plus and phenotypic Löwenstein-Jensen method for isoniazid or rifampicin drug susceptibility tests in tuberculosis patients.

Background: The widespread use of molecular, genotypic drug susceptibility tests (DSTs) for antituberculosis (anti-TB) drugs has led to the dilemma of interpreting discordant results between genotypic and conventional, phenotypic DSTs. We investigated the clinical characteristics, including treatment patterns and outcomes, of TB patients with a genotype-phenotype discrepancy in susceptibility to isoniazid (INH) or rifampicin (RIF).
Methods: We retrospectively reviewed the medical records of TB patients who had results for 2 DSTs (genotypic method, MTBDR plus test for INH and RIF, and phenotypic method) treated between August 2010 and October 2016 in a tertiary university hospital.
Results: Among 1,069 TB patients, 63 (5.9%) had discrepant results for the 2 DSTs. Of the 57 multidrug-resistant (MDR) TB cases diagnosed by either DST, 18 (31.6%) showed discordant results for INH or RIF. The most frequent pattern of discordance was genotypic susceptibility with phenotypic resistance to INH. RIF-discordant subjects with genotypic resistance were more likely to have been exposed previously to anti-TB drugs and to have an MDR TB diagnosis and concurrent INH resistance. Forty-five of the 54 patients managed in our hospital (83.3%) had a favorable outcome with a mean treatment duration of 14.0 months. Ten of the 16 INH-discrepant patients with a genotypic mutation continued taking INH, but more than half patients in the RIF-discrepant group (8/14) with a genotypic mutation discontinued taking RIF.
Conclusions: Despite the low frequency, discordant results were obtained between the genotypic and phenotypic DSTs for INH or RIF, especially for patients with MDR TB or INH resistance. Furthermore, it seemed that RIF discrepancy with a genotypic mutation might have a greater impact on the clinical outcome than INH discrepancy.
Competing Interests: Conflicts of Interest: The authors have no conflicts of interest to declare.