Determination of the susceptibility of Mycobacterium tuberculosis to pyrazinamide in liquid and solid media assessed by a colorimetric nitrate reductase assay.

OBJECTIVES

The aim of this study was to develop and evaluate two colorimetric nitrate reductase-based antibiotic susceptibility (CONRAS) tests, the CONRAS-liquid test and the CONRAS-LJ test, to enable susceptibility testing of Mycobacterium tuberculosis to pyrazinamide. To enhance the growth of M. tuberculosis in 7H9 broth with acid pH (6.0), the effect of three potential growth-promoting substances (reconstitution fluid, fastidious organism supplement and epidermal growth factor) was evaluated.

METHODS

Seventy-five M. tuberculosis strains were tested for susceptibility to pyrazinamide in the CONRAS-liquid test performed in Middlebrook 7H9 broth, and 77 M. tuberculosis strains were tested for susceptibility to pyrazinamide in the CONRAS-LJ test performed on Löwenstein-Jensen medium. The BACTEC 460TB system, sequencing of the pncA gene and Wayne's assay were used as reference tests. Growth of 10 M. tuberculosis strains in conventional 7H9 broth and in acid 7H9 broth with and without growth-promoting substances added was evaluated by the nitrate reductase assay.

RESULTS

By using the BACTEC 460TB system as the reference test, the sensitivity and specificity of the CONRAS-liquid test (100 mg/L pyrazinamide) tested on 75 M. tuberculosis strains were 87.5% and 83.7%, respectively. The corresponding sensitivity and specificity of the CONRAS-LJ test (1200 mg/L pyrazinamide) tested on 77 M. tuberculosis strains were 100% and 84.1%, respectively. The mean turn around time of the CONRAS-liquid test was significantly shorter than that of the CONRAS-LJ test (mean, 9.4 and 14.5 days, respectively; P < 0.001). In addition, no effect of growth-promoting substances on the growth of M. tuberculosis in a broth with acid pH was seen.

CONCLUSIONS

The CONRAS tests were rapid, cheap and easy to perform and interpret. Both tests should be evaluated on extended strain batteries in multicentre studies before they can be considered for use in susceptibility testing of M. tuberculosis to pyrazinamide in resource-limited settings.