Molecular detection of Mycobacterium tuberculosis: impact on patient care

Abstract

Background: Nucleic acid amplification technologies such as PCR are revolutionizing the detection of infectious pathogens such as tuberculosis (TB). Amplification technology offers the potential for the diagnosis of TB in a few hours with a high degree of sensitivity and specificity. However, molecular assays neither replace nor reduce the need for conventional smear and culture, speciation, and antibiotic sensitivity assays.

Methods: We undertook prospective studies of sputum samples to assess the performance of two PCR-based assays for the detection of TB as well as the impact of more rapid availability of test results on patient care.

Results: The sensitivity of both the in-house and Amplicor PCR assays was 100% for smear-positive sputa. For smear-negative sputa (two sputum samples collected during the first 24 h of hospitalization), the sensitivity was 85% for our in-house PCR assay and 74% for the Roche PCR assay. Approximately 10% of the smear- and culture-negative sputa yielded positive PCR results; however, more than one-half of these were positive with both the in-house and Amplicor assays, suggesting the presence of TB DNA or organisms. Several of these came from patients whose other samples grew Mycobacterium tuberculosis during the same admission, and others came from patients who had previously treated TB. Overall, the specificities of the in-house and Amplicor PCR assays in smear-negative patients were 86% and 93%, respectively.

Conclusions: Molecular detection of slow-growing pathogens such as M. tuberculosis have the potential to improve clinical care through a dramatic reduction in the time required for detection and may provide substantial savings in the overall cost of care of a patient compared with conventional smear, culture, and speciation alone, despite the fact that conventional assays must still be performed for speciation of nontuberculous mycobacteria and for full assessment of antibiotic sensitivity.