Practical strategies for performance optimization of the enhanced gen-probe amplified mycobacterium tuberculosis direct test.

The enhanced Gen-Probe Amplified Mycobacterium Tuberculosis Direct (MTD) test was evaluated using a combined set of 338 acid-fast smear-positive and smear-negative, respiratory and nonrespiratory clinical specimens received by the Massachusetts State Tuberculosis Laboratory from September 1999 through March 2002. Microbiological culture was used as the reference method; therefore, the sensitivity and specificity of the MTD test were calculated for culture-positive specimens only. The initial assessment indicated that the overall sensitivity, specificity, and positive and negative predictive values of the MTD test for all specimens grouped together were 62, 98, 99, and 68%, respectively. A detailed discrepancy analysis revealed that two major factors causing negative MTD results in specimens that were culture positive for M. tuberculosis complex were patient treatment with antituberculosis drugs prior to testing and the presence of inhibitory substances in the specimen. Based on these findings, a protocol for optimizing MTD test performance in this setting is proposed in which (i) specimens from patients taking antituberculosis medications are excluded from testing and (ii) all initially MTD-negative or MTD-equivocal specimens are subjected to testing for inhibitors. If this strategy was followed, the MTD test sensitivity would be at least 91%, a significant improvement over the initial sensitivity of 62%. Accordingly, the negative predictive value would increase from 68 to 91%.