Detection of Mycobacterium tuberculosis in BACTEC 12B broth cultures by the Roche Amplicor PCR assay

Abstract

To evaluate the ability of the Amplicor MTB Assay to detect Mycobacterium tuberculosis complex (MTBC) organisms in BACTEC 12B broth cultures, 249 cultures with a growth index (GI) of > or = 20 from 160 patients were tested retrospectively. Specimens were processed by standard methods, and then BACTEC 12B vials and Middlebrook 7H11/7H115 plates were inoculated, incubated, and interpreted in accordance with the manufacturer's instructions and laboratory protocol. From 12B vials with a GI of > or = 20, and aliquot of broth was removed and frozen at -20 degrees C until assayed by PCR. PCR results were compared to those obtained by the usual laboratory protocol, whereby MTBC organisms were identified by a DNA probe assay performed on broth from 12B vials with a GI or > or = 300 or on colonies from solid medium. Of the 249 broth cultures evaluated, 142 contained mycobacteria, including 44 that contained MTBC organisms. Of these 44 cultures, 41 were PCR positive; the 3 that were PCR negative were blood specimens collected in an Isolator tube. All 98 cultures with nontuberculous mycobacteria and the 107 that did not contain mycobacteria were PCR negative. Thus, the sensitivity and specificity of PCR were 93 and 100%, respectively. For those culture sin which MTBC organisms were identified by both the DNA probe and PCR assays, the mean time from specimen inoculation to detection and identification of MTBC organisms was 16 (range, 4 to 26) days for the PCR and 28 (range, 13 to 43) days for the DNA probe assay (P < 0.0001). In summary, PCR is a rapid, reliable method for detection of MTBC organisms in BACTEC 12B broth cultures with a GI of > or = 20.