Detection and identification of Mycobacterium tuberculosis directly from sputum sediments by Amplicor PCR

Abstract

Sputum specimens received for the diagnosis of tuberculosis or other mycobacterial infections were tested by a PCR-based assay and culture techniques. Results of the PCR assay (Amplicor Mycobacterium tuberculosis Test) were compared with results of standard culture techniques with cultures held for 6 weeks. One thousand nine specimens were included: 301 retrospective specimens (frozen at -70 degrees C and later tested by PCR) and 708 prospective specimens (tested within 1 day of processing). One hundred sixty-two (16%) of the specimens were culture positive for M. tuberculosis; 83 (51%) of these were also fluorochrome stain positive. The sensitivity and specificity of the Amplicor PCR compared with those of culture were 83% (134 of 162 specimens) and 97% (800 of 827 specimens), respectively. The sensitivity for fluorochrome stain-positive specimens was 99%, and that for fluorochrome stain-negative specimens was 66%. The great majority of the 28 PCR-negative, culture-positive specimens were low positives; 27 were smear negative and 19 contained < 100 CFU of M. tuberculosis per ml. The 27 PCR-positive, culture-negative specimens included 24 that were positive by repeat testing by alternate primer PCR and were from patients with tuberculosis on antimicrobial therapy. With these considered culture misses, the final sensitivities of PCR and culture were 85, and 87%, respectively, while the specificities were 99.6 and 100%, respectively. After normal laboratory processing of sputum specimens, the Amplicor PCR assay can be completed in 8 h. Thus, it is possible to have results available within 10 h of specimen submission.