Evaluation of mycobacteria growth indicator tube for recovery and drug susceptibility testing of Mycobacterium tuberculosis isolates from respiratory specimens.

The new BBL mycobacteria growth indicator tube (MGIT) was evaluated for its ability to detect mycobacteria directly from patient specimens and to determine the drug susceptibility of Mycobacterium tuberculosis isolates. A total of 85 respiratory specimens were tested. Specimens were digested, concentrated, examined microscopically for acid-fast bacilli, and inoculated into MGITs and onto Lowenstein-Jensen slants by standard procedures. The tubes were incubated at 37 degrees C and were examined daily for fluorescence to 365-nm UV light. All 25 specimens smear positive for acid-fast bacilli were tested for drug susceptibility in MGITs containing 1.0 mu g of rifampin per ml, 0.1 mu g of isoniazid per ml, 2.0 mu g of streptomycin per ml, and 2.0 mu g of ofloxacin per ml. These results were compared with those obtained by testing the same M. tuberculosis isolates by the indirect proportion method at drug concentrations of 4.0 mu g of rifampin per ml, 0.2 mu g of isoniazid per ml, 2.0 mu g of ethambutol per ml. 4.0 mu g of streptomycin per ml, and 2.0 mu g of ofloxacin per ml. No significant difference in the sensitivity of detection of M. tuberculosis isolates was found between the two methods. However, the time to detection was significantly shorter in MGITs. Drug susceptibility test results for M. tuberculosis isolates by the two methods demonstrated an excellent correlation. The mean time to reporting of drug susceptibility results was 5 days for MGITs versus 16 days for Lowenstein-Jensen slants. The results of this preliminary study indicate that the MGIT system appears to have potential for routine use in mycobacteriology for both the detection and the drug susceptibility testing of M. tuberculosis isolates. However, it is important to emphasize that simple nonautomated equipment should be developed to improve the accuracy of fluorescence detection.