Setting: National Tuberculosis Clinical Laboratory, China.
Objective: To evaluate the accuracy and feasibility of the MYCOTB MIC plate in anti-tuberculosis drug susceptibility testing.
Design: MYCOTB testing of Mycobacterium tuberculosis isolates, the Löwenstein-Jensen (LJ) proportion method and the resurazine microtitre assay (REMA), which is extensively used for in-house assay for minimal inhibition concentration (MIC) testing, were performed simultaneously for comparison. A total of 126 clinical isolates were tested using both MYCOTB and the LJ proportion method against 12 anti-tuberculosis drugs; 80 were also tested using REMA.
Results: Categorical agreement between MYCOTB and the LJ proportion method was 99.2% for rifampicin, ofloxacin, amikacin, kanamycin and cycloserine, and 98.4% for isoniazid and para-aminosalicylic acid; ethambutol (EMB) had the lowest agreement (86.5%). The overall categorical agreement between MYCOTB and REMA ranged between 98.8% and 100%. MYCOTB outcomes, interpreted on day 10 and 21, were stable for all drugs except EMB.
Conclusion: MYCOTB is a rapid, convenient, quantitative and accurate method for testing both first- and second-line anti-tuberculosis drugs.