Diagnostic performance of amplified Mycobacterium tuberculosis direct test with cerebrospinal fluid, other nonrespiratory, and respiratory specimens.

The Gen-Probe Amplified Mycobacterium tuberculosis Direct Test (MTD) was adapted to be used for cerebrospinal fluid (CSF) and a large variety of other nonrespiratory specimens. Standardized with artificially spiked dilution series of CSF, the modified MTD procedure consists of (i) increasing the amount of sample 10-fold, (ii) pretreating the specimen with a detergent, and (iii) increasing the amplification time from 2 to 3 h. Performance of MTD in a clinical mycobacteriology laboratory was tested over an extended period of time, involving a total of 322 nonrespiratory as well as 1,117 respiratory specimens from 998 patients. Results from MTD were compared with those from microscopy, culture, analysis of tuberculostearic acid by gas-liquid chromatography-mass spectrometry (CSF only), and the final clinical diagnosis. When MTD results were compared with resolved data, the sensitivity, specificity, and positive and negative predictive values for MTD were 93.1, 97.7, 90.0, and 98.5%, respectively, for nonrespiratory specimens and 86.6, 96.4, 76.8, and 98.1%, respectively, for respiratory specimens. Our data demonstrate that (i) MTD is a robust, highly sensitive and specific technique for the rapid detection of M. tuberculosis complex in all types of clinical specimens, (ii) there was no statistically significant difference (P > 0.005) in sensitivity and specificity for nonrespiratory compared with respiratory specimens, and (iii) repeating all MTDs which yield a result between 30,000 and 200,000 relative light units would help prevent a large number of false positives and, thus, enhance test specificity.