Molecular diagnosis of tuberculous meningitis a 4year experience
Abstract number: O238
Mendes A., Almeida S., Fernandes S., Rodrigues K., Santos A., Ramos H.
Objectives: Tuberculous meningitis (TBM) is the most severe form of extrapulmonary tuberculosis, with high mortality rates and serious long-term consequences. Accurate and early diagnostic confirmation is essential in patient management. The purposes of this study was to, retrospectively, evaluate the performance of a molecular test for the detection of Mycobacterium tuberculosis complex in cerebrospinal fluid specimens (CSF), compared with culture and microscopy, and correlate the results with clinical findings.
Methods: Between November 2005 and September 2009, 118 CSF specimens were sent to the Molecular Biology Unit of Centro Hospitalar do Porto, with requests for M. tuberculosis molecular detection (MTD, GenProbe), with mycobacterial culture (MGITTM and Lowenstein-Jensen) and microscopy examination performed in the Microbiology laboratory. MTD test was performed as described by manufacturer, but after mechanical cell lysis, nucleic acid purification was carried out in EZ1 BioRobot (Qiagen). Presence of amplification inhibition was verified in every sample. MTB diagnosis was based in laboratory results and clinical criteria, such as patient response to anti-bacillary therapy.
Results: 118 samples from 107 patients were studied. Molecular detection of M. tuberculosis complex was positive in 5 cases, which were confirmed by cultural methods. Microscopy revealed acid-fast bacilli in one sample. MTB diagnosis (laboratory and clinical criteria) was established in 17 cases, 12 being presumptive. All 17 patients showed altered cytology with pleocytosis. Protein level was elevated in 16 samples (0.5510.5 g/L). Compared with MTB diagnosis, MTD showed a sensitivity value of 30%, specificity of 100%, positive predictive value (PPV) of 100% and a negative predictive value (NPV) of 89%.
Conclusion: Our results suggest the potential role of molecular methods in confirming diagnosis of MTB, with short turnaround time results (2.5 h), high PPV, and similar sensitivity as culture (30%). Results must be interpreted in parallel with clinical findings and conventional microbiology methods. The use of molecular tests should not exclude a diagnosis of MTB (NPV-89%). More studies would help improve the adequacy of these methods.
|Session name:||Abstracts 20th European Congress of Clinical Microbiology and Infectious Diseases|
|Location:||Vienna, Austria, 10 - 13 April 2010|
|Back to top|