Abstract:

Objective To explore the value of Micropore-plate method (Microdrug sensitivity, MicroDST) test in detecting sensitivity of first-line anti-tuberculosis drugs. Methods A total of 1086 specimens cultured positively by the MGIT 960 method were collected from Xi’an Chest Hospital from January to June 2019, and 331 bacterial liquids were selected which were identified as Mycobacterium tuberculosis and were tested for drug snensitivity by MicroDST and MGIT 960 method. Results of the two methods were analyzed and the detection efficiency of was evaluated with the gold standard-MGIT 960 test. Melting curve method were used to verify the resistant genotypes when test results of bacterial liquids were different. Results Based on MGIT 960 test, the sensitivity, specificity and Kappa values of MicroDST for streptomycin, isoniazid, rifampicin and ethambutol were 88.7% (63/71), 100.0% (260/260), 0.93; 93.9% (77/82), 98.8% (246/249), 0.94; 93.8% (45/48), 99.6% (282/283), 0.95; and 66.7% (14/21), 99.0% (307/310), 0.72 respectively. Among the inconsistent results of the four drug resistance tests using the MicroDST method and the MGIT 960 method, there were 10 for streptomycin, 8 for isoniazid, 6 for rifampicin, and 16 for ethambutol. In the 10 specimens (3.0%, 10/331) which were medium sensitive for MicroDST and resistant for MGIT 960 test, drug resistance gene were mutant (2 for streptomycin, 2 for rifampicin, and 6 for ethambutol). Only one specimen (0.3%, 1/331) was medium sensitive for ethambutol in MicroDST and sensitive in MGIT 960 test, the drug resistance gene was mutant. In the 23 specimens (6.9%, 23/331) which were sensitive in MicroDST and resistant in MGIT 960 test, there were 10 specimens which drug resistance gene were wild type (6 for streptomycin and 4 for isoniazid) and 13 specimens which drug resistance gene were mutant (2 for streptomycin, 1 for isoniazid, 3 for rifampicin, and 7 for ethambutol). In the 6 specimens (1.8%, 6/331) which were resistant in MicroDST and sensitive in MGIT 960 test, there were 3 specimens which drug resistance gene were wild type (1 for isoniazid, 1 for rifampicin, and 1 for ethambutol, respectively) and 3 specimen which drug resistance gene were mutant (2 for isoniazid and 1 for ethambutol). Conclusion The MicroDST method is simple and convenient for the detection of first-line anti-tuberculosis drugs. It is highly consistent with the MGIT 960 method and can detect drug resistance at low concentrations, which can provide a reference for clinical drug dosage selection.

Key words: Mycobacterium tuberculosis, Tuberculosis,multidrug-resistant, Microbial sensitivity tests, Comparative study