Comparison of Amplicor, in-house PCR, and conventional culture for detection of Mycobacterium tuberculosis in clinical samples.

Five hundred four clinical specimens (337 sputum and 167 bronchial samples) from 340 patients were tested for the presence of M. tuberculosis complex by the Amplicor M. tuberculosis test and by an in-house PCR. The results were compared with those obtained by conventional culture and by direct microscopy. Thirty specimens (from 14 patients) were positive by in-house PCR, 25 (from 13 patients) were positive by the Amplicor M. tuberculosis test, and 24 (from 10 patients) were positive by culture. Cultures from 16 specimens were contaminated with other bacteria. Strong inhibition of in-house PCR was found with three samples. After discordancy analyses, with clinical data as supportive evidence for tuberculosis, 27 true-positive and 458 true-negative samples were defined. On the basis of these figures, the sensitivities of the Amplicor M. tuberculosis test, in-house PCR, culture, and microscopy were 70.4, 92.6, 88.9, and 52.4%, respectively. The specificities of all four tests were higher than 98%. The good performance of the in-house PCR for detection of M. tuberculosis makes it a very useful additional tool in M. tuberculosis diagnostics. In contrast, the Amplicor test needs to be improved. Twenty-three of the Amplicor-negative samples were further tested for inhibition of the Amplicor system by retesting the DNA extracts after the addition of M. tuberculosis DNA. In 15 of these samples, 5 true positives and 10 true negatives, inhibition of the Amplicor test was demonstrated. This might explain the lack of sensitivity of the Amplicor test. If the inhibition problem can be solved, the Amplicor M. tuberculosis test, which is already rapid, very user-friendly, and reasonably priced, may certainly become very useful in microbiological laboratories.