Comparison of Culture and PCR Methods for Diagnosis of Mycobacterium tuberculosis in Different Clinical Specimens

BACKGROUND Tuberculosis remains a global epidemic, especially in developing countries, including Iran. Rapid diagnosis of active Mycobacterium tuberculosis infection plays a critical role in controlling the spread of tuberculosis. Conventional methods may take up to several weeks or longer to produce results. In addition to multiplicity of steps involved in conventional detection, including isolation, identification and drug susceptibility testing, the slow growth rate of M. tuberculosis is also responsible for this lengthy time. OBJECTIVES The aim of this study was to compare the polymerase chain reaction (PCR) and culture methods for the detection of M. tuberculosis in different clinical specimens. MATERIALS AND METHODS This study was performed on different samples (urine, gastric aspirate, bronchoalveolar lavage, pleural fluid, cerebrospinal fluid, ascetic fluid and joint fluid specimens) of tuberculosis suspected patients. M. tuberculosis DNA was extracted directly from different samples using two different protocols. Next, PCR was performed using three sets of specific primers to detect members of Mycobacterium genus, M. tuberculosis complex and non-tuberculosis Mycobacteria. The results were then compared with that of the culture method, which is considered as the gold standard method. RESULTS The concordance rate between the three sets of primers was calculated and IS6110/buffer PCR method showed good agreement with the LJ culture method (κ = 0.627, P < 0.0001). The sensitivity of IS6110/buffer PCR was 58.33%, with specificity of 77.78%; the positive and negative predictive values were 100% and 78.26%, respectively. Buffer method for DNA extraction was proved to give a higher accuracy to PCR in comparison with the boiling method. CONCLUSIONS PCR method is a valuable, cost-effective and alternative tool for quick diagnosis of active tuberculosis in different clinical specimens.