Objectives: Tuberculosis, a reemergent killer, is threatening to assume serious proportions all over the world, particularly in view of the AIDS pandemic. The detection of mycobacterial DNA by polymerase chain reaction (PCR) in clinical samples is a promising approach for the rapid diagnosis of tuberculous infections. The aims of this study were to evaluate PCR for detection of Mycobacterium tuberculosis in pleural fluids and to correlate the results with adenosine deaminase activity (ADA) estimation and acid-fast bacilli (AFB) screening.
Methods: The sensitivity and specificity of PCR in detection of mycobacterial DNA in 20 samples of tuberculous pleural effusion were evaluated using 40 samples of nontubercular pleural effusion as controls. The results were correlated with the ADA in all 60 pleural fluids. In addition, AFB detection by Ziehl-Neelsen staining on cytospin smears of all pleural fluids was also compared.
Results: Of the 20 samples of tuberculous pleural effusion, mycobacterium could be detected by AFB staining in 4 samples. Fourteen samples were PCR positive. None of the samples from the control group were AFB or PCR positive. The sensitivity of PCR, therefore, was 70.0% with specificity of 100% (positive predictive value, 100%; negative predictive value, 86.95%). The sensitivity of AFB screening was at best 20%. The mean of ADA values in tubercular pleural effusions was 63.21 U/L (SD, 33.01), and the mean in the control samples was 51.1 U/L (SD, 29.71). Taking a cut-off value of 50 U/L, both the sensitivity and specificity of ADA estimation in diagnosing tuberculosis were only 55%.
Conclusion: PCR represents a rapid and sensitive method for the detection of mycobacterial DNA in tuberculous pleural effusions. AFB screening has low sensitivity, and ADA estimation has both low sensitivity and specificity. Therefore, when the clinical suspicion is high and smear result is negative, but the signs and symptoms of M tuberculosis are apparent, PCR is the method of choice for identifying the infection.