Comparison of three methods for susceptibility testing of Mycobacterium avium subsp. paratuberculosis to 11 antimicrobial drugs

Objectives To evaluate the BACTECTM MGITTM 960/MGIT Para TB (MGIT) system for drug susceptibility testing of Mycobacterium avium subsp. paratuberculosis (MAP), a pathogen implicated in some forms of Crohn's disease. Methods MICs of 11 drugs for 10 MAP strains were determined using the MGIT system, the BACTECTM460TB system (BACTEC) and conventional agar dilution methods. Results MICs determined by MGIT methods showed 80%–100% agreement (±1 log2 dilution) with those determined by the BACTEC and agar dilution methods for ciprofloxacin, levofloxacin, azithromycin and clofazimine. The MGIT and BACTEC methods showed 70%, 80% and 90% agreement (±1 log2 dilution) for MICs of ethambutol, rifabutin and rifampicin; agreement for all drugs increased to 100% at 2 log2 dilution differences. For clarithromycin, the MGIT method had greater agreement with the agar dilution method (70% at the same dilution) than the BACTEC method (60% at ±1 log2 dilution); agreement increased to 100% at ±2 log2 dilutions in both cases. The MGIT and agar dilution methods agreed 60% and 100% for amikacin MICs at ±1 log2 dilution and ±2 log2 dilutions, respectively. By all methods MICs were higher than achievable serum concentrations for isoniazid and dapsone. There was 100% agreement between all three methods for azithromycin, clarithromycin and ciprofloxacin, and 80% agreement for rifampicin using published MIC thresholds available for M. avium complex strains. Conclusions This study shows that the MGIT system can be used for rapid and reliable drug susceptibility testing of MAP.