ABSTRACT
Introduction: Nucleic acid-based amplification tests are currently licensed only for the detection of Mycobacterium tuberculosis in pulmonary specimens. There are insufficient data for extrapulmonary specimens. The aim of this study is to investigate the diagnostic value of these investigations in tuberculous meningitis. Materials and Methods: We performed a prospective study using a commercial ligase chain reaction DNA amplification technique [Ligase chain result (LCx) M. tuberculosis; Abbott Laboratories, Abbott Park, IL, USA] on cerebrospinal fluid (CSF) to diagnose tuberculous meningitis and compared the results with standard microbiological data. Conflicting cases were resolved according to the final clinical diagnosis. A total of 54 CSF specimens from 54 patients were tested. Results: Six (11.1%) specimens were culture-positive for M. tuberculosis; of these, only 1 (1.9%) was smear-positive. The sensitivity, specificity, positive predictive value and negative predictive value of the LCx assay, compared with culture results, were 66.7%, 100%, 100% and 96%, respectively. After resolution according to clinical data, the sensitivity, specificity, positive predictive value and negative predictive value were 33.3%, 100%, 100% and 75%, respectively. All controls had negative LCx results. There were no false positives. Conclusion: LCx assay is highly specific and complements conventional laboratory diagnostic methods (CSF smear and culture) in the diagnosis of tuberculous meningitis. In the appropriate clinical context, a positive result is strongly suggestive of tuberculosis and could enable antituberculous treatment to be started immediately.
Tuberculous meningitis (TBM) is a serious disease with 100% mortality in untreated patients. In patients in whom treatment is delayed, severe neurological deficits result.1 Hence, rapid diagnosis of TBM is important but difficult by conventional diagnostic methods. Direct smears of cerebrospinal fluid (CSF) for acid-fast bacilli (AFB), although virtually diagnostic, are usually positive in fewer than 10% of cases of TBM.2,3 Culture on solid media takes up to 8 weeks.4 The radiometric BACTEC 460 TB system (Becton Dickinson Diagnostic Instrument Systems, Sparks, MD, USA) requires an average of 13 to 15 days to detect positive specimens.5
This article is available only as a PDF. Please click on “Download PDF” to view the full article.
