Optimization of the microscopic observation drug susceptibility assay for four first-line drugs using Mycobacterium tuberculosis reference strains and clinical isolates

Abstract

ObjectiveThe aim of this study is to determine the appropriate cut-off value and turnaround time of the microscopic observation drug susceptibility assay (MODS) for isoniazid (INH), rifampicin (RMP), streptomycin (STR), and ethambutol (EMB). DesignA total of 39 Mycobacterium tuberculosis strains with confirmed drug susceptibility (reference strains) were tested with a range of drug concentrations to determine the optimal cut-off values for INH, RMP, STR, and EMB by MODS. Standard drug susceptibility testing (DST) results were evaluated relative to the Löwenstein–Jensen (L–J) proportion method. Following which, the performance of MODS was evaluated again using 36 sputum samples from patients with tuberculosis (TB) using the cut-off values determined in the aforementioned process. ResultsWith 39 reference strains, DST identified the following cut-off values: 0.8μg/ml INH (sensitivity, 96.0%; specificity, 92.9%), 2.0μg/ml RMP (sensitivity, 100%; specificity, 95.5%), 4.0μg/ml STR (sensitivity, 90.5%; specificity, 93.8%), and 4.0μg/ml EMB (sensitivity, 100%; specificity, 91.7%). When these cut-off values were used to analyze the 36 clinical isolates, the sensitivity and specificity of MODS were 100% and 93.1% for INH, 100% and 93.8% for RMP, 87.5% and 96.4% for STR, and 100% and 88.2% for EMB, respectively. The turnaround time for these clinical specimens was 9.0days by MODS (95% CI: 5.3–12.7), compared with 11.7days (95% CI: 9.5–13.9) for smear negative specimens. ConclusionOur study identified the optimal cut-off values of the four first-line drugs for MODS based on a wide concentration range. With the optimal cut-off values determined in this study, MODS showed high discriminatory efficiency for DST. This study also demonstrated that MODS is useful for rapid diagnosis of drug-resistant TB even for a smear negative specimen, despite the fact that it generally uses smear positive specimens as direct DST.