DNA sequencing of rpoB and culture-based drug susceptibility results were evaluated for samples referred for confirmation of rifampin resistance detected by the Cepheid Xpert MTB/RIF assay. Silent mutations and mutations associated with low-level resistance were found in the study population. These data support CDC recommendations to confirm Xpert rifampin resistance results.R apid diagnosis and effective treatment are two of the most important strategies in a tuberculosis (TB) control program to prevent ongoing transmission of disease and to improve patient outcomes (1). To ensure an effective treatment regimen, drug susceptibility testing (DST) must be performed. Culture-based methods can take 4 to 12 weeks and are considered the "gold standard" for DST of Mycobacterium tuberculosis complex (MTC). To provide more rapid DST results, the Centers for Disease Control and Prevention (CDC) implemented a clinical laboratory service, Molecular Detection of Drug Resistance (MDDR), for U.S. Public Health TB programs in September 2009. MDDR uses PCR to amplify targeted genetic loci and then DNA sequencing to detect mutations associated with resistance to four first-line drugs, rifampin (RIF), isoniazid (INH), pyrazinamide (PZA), and ethambutol (EMB), and 4 second-line drugs, ofloxacin (OFX), amikacin (AMK), capreomycin (CAP), and kanamycin (KAN) (2). Culture-based DST by the agar proportion method and MGIT for PZA, is performed concurrently on all samples received for MDDR. Culture-based DST is necessary to complement molecular results because the clinical relevance of some mutations is unknown, and not all mechanisms of resistance are understood (3). However, culture-based DST for RIF is imperfect, and DNA sequencing may yield more information in some situations (4-6). More than 95% of RIF-resistant (RIF r ) strains contain a mutation in the RIF resistance-determining region (RRDR) of rpoB. Previous studies revealed mutations in the RRDR (e.g., 511Pro, 516Tyr, 526Asn, 526Leu, and 533Pro) that are "disputed" or associated with highly discordant results among culture-based DST methods because they yield low-level RIF r that may not be detected by some methods. TB cases that are caused by strains exhibiting these mutations may not respond well to a rifampin-based treatment regimen (4-6). Some strains may harbor silent mutations in the RRDR that do not result in an amino acid change and do not confer resistance (7).In July 2013, the Cepheid Xpert MTB/RIF assay (Xpert) received FDA market authorization for the primary identification of MTC and the detection of RIF r by molecular analysis. This assay can be performed on raw or concentrated sputum sediments using the fully automated GeneXpert Instrument and provides rapid results weeks earlier than culture-based methods (8). However, some experts have recommended that RIF r detected by Xpert be confirmed by DNA sequencing due to the potential low positive predictive value for detection of RIF r , attributable to the low prevalence of drug resistance among U.S. TB cases and bec...

Ethambutol (EMB) is used as a part of drug regimens for treatment of tuberculosis (TB). Susceptibility of Mycobacterium tuberculosis complex (MTBC) isolates to EMB can be discerned by DNA sequencing to detect mutations in the embB gene associated with resistance. US Public Health Laboratories (PHL) primarily use growth-based drug susceptibility test (DST) methods to determine EMB resistance. The Centers for Disease Control and Prevention (CDC) provides a service for molecular detection of drug resistance (MDDR) by DNA sequencing and concurrent growth-based DST using agar proportion. PHL and CDC test results were compared for 211 MTBC samples submitted to CDC from September 2009 through February 2011. Concordance between growth-based DST results from PHL and CDC was 88.2%. A growth-based comparison of 39 samples, where an embB mutation associated with EMB resistance was detected, revealed a higher percentage of EMB resistance by CDC (84.6%) than by PHL (59.0%) which was significant (P value = 0.002). Discordance between all growth-based test results from PHL and CDC was also significant (P value = 0.003). Most discordance was linked to false susceptibility using the BACTEC™ MGIT™ 960 (MGIT) growth-based system. Our analysis supports coalescing growth-based and molecular results for an informed interpretation of potential EMB resistance.