Details
Zusammenfassung: <jats:p>A total of 294 clinical respiratory specimens, including 75 with culture-positive results, were tested for the presence of Mycobacterium tuberculosis by strand displacement amplification (SDA) of DNA. A region of the IS6110 insertion element and an internal control sequence were amplified and then detected by a chemiluminescence assay. Receiver operator-characteristic curves were used to evaluate three methods for declaring specimens positive for M. tuberculosis. By the preferred method, SDA chemiluminescence results were converted to theoretical numbers of M. tuberculosis organisms. A positive threshold (PT) value, above which 95% of the SDA results were judged to be M. tuberculosis positive (sensitivity = 95%), was found to be 2.4 M. tuberculosis organisms per SDA reaction. The analogous PT value for 95% sensitivity on smear-positive specimens was 3.6 M. tuberculosis organisms per reaction. The PT of 2.4 M. tuberculosis organisms per reaction detected 100% of culture-positive, smear-positive specimens (sensitivity = 100%), while 95% sensitivity was achieved with a PT of 15.5 M. tuberculosis organisms per reaction. Specificities, which were calculated with respect to culture- and smear-negative specimens, ranged from 96% at a PT of 15.5 M. tuberculosis organisms to 84% at a PT of 2.4 M. tuberculosis organisms per reaction. The M. tuberculosis-negative specimens were also segregated according to whether the patients received antituberculosis chemotherapy. SDA specificity ranged from 90% (PT = 2.4 M. tuberculosis organisms) to 98% (PT = 15.5 M. tuberculosis organisms) for the M. tuberculosis-negative specimens from patients who had not received chemotherapy. SDA specificity in the M. tuberculosis-negative specimens from patients who received chemotherapy was lower (85 to 94%). This study represents the first large-scale demonstration of M. tuberculosis detection in clinical sputum specimens by isothermal DNA amplification with SDA.</jats:p>
Umfang: 860-865
ISSN: 1098-660X
0095-1137
DOI: 10.1128/jcm.34.4.860-865.1996