[Show abstract][Hide abstract] ABSTRACT: Nucleic acid amplification techniques are being used increasingly in diagnosing tuberculosis. In developing countries clinical samples are often stored for subsequent analysis since molecular tests are conducted at only a limited number of laboratories. This study was conducted to assess the speed at which mycobacteria undergo autolysis and free DNA is detected in the supernatant during low-temperature storage.
Eighty-seven smear positive sputa from tuberculosis patients were analysed immediately and after storage at -20 degrees C. Timelines of 1 and 2 months were selected to assess the maximum extent of DNA loss that occurred during storage. All samples remained PCR- and smear-positive at 1 month and only 1 sample turned negative after 2 months. Bacterial lysis in the specimens was demonstrated by PCR analysis of supernatant fractions; 53% of the freshly analysed samples contained mycobacterial DNA in supernatants. PCR positivity increased significantly during storage (to 69% and 77% after 1 and 2 months of storage, respectively, P < 0.0001). Storage-associated bacterial lysis was accompanied by a decrease in smear grade status in 28 of 87 samples (P < 0.0001 after 2 months of storage) and a significant storage-associated reduction in bacterial numbers in the remaining samples.
We conclude that (i) freshly isolated sputum contains both intact and lysed mycobacteria, (ii) lysis increased during storage and (iii) supernatant fractions routinely discarded during sample processing contain mycobacterial DNA. We propose that supernatant is a valuable sample for PCR for both fresh and stored specimens, particularly those with a low bacterial load in addition to conventional sediment.
[Show abstract][Hide abstract] ABSTRACT: This study was designed to assess the vital issue of gene target length and PCR assay performance in relation to the detection of Mycobacterium tuberculosis in clinical specimens. Two PCR assays that amplify fragments of varying lengths from the devR gene of M. tuberculosis were evaluated. Using M. tuberculosis DNA the 'short-length' PCR assay detected 250-500 genome equivalents vs. 500-1,000 genome equivalents by the 'long-length' assay. In comparison to a highly sensitive smear microscopy test (universal sample processing smear), the sensitivity of the 'short-length' assay was 97.8% vs. 69.9% of the 'long-length' assay in sputum specimens (n=506) from patients being evaluated for a possible diagnosis of tuberculosis. The 27.9% absolute increase in sensitivity was statistically significant (P<0.001). Our results indicate that in a clinical setting when all other conditions are equal, the amplification of a shorter gene fragment of devR increases the sensitivity and efficiency of the PCR assay in spite of using a single copy gene as target.