ELISA in tuberculous meningitis: using two mycobacterial antigens does not improve the diagnostic yield compared to either antigen alone

Antibody detection tests for early diagnosis in tuberculous meningitis

BACKGROUND

Tuberculous meningitis (TBM) is the most severe form of tuberculosis. Microbiological confirmation is rare and treatment is often delayed. Early diagnosis and immediate initiation of treatment are essential for effective TBM control. A systematic review was performed in this study to assess the diagnostic accuracy of detecting antibodies against Mycobacterium tuberculosis in the cerebrospinal fluid (CSF), according to standard methods. Test performance was summarized using a bivariate random-effects meta-analysis.

METHODS

Studies were identified by a search of the literature, up to July 25, 2015, in the EMBASE and MEDLINE databases via Ovid SP and PubMed. The Cochrane Library was also searched for original, peer-reviewed molecular epidemiology studies that reported the diagnosis of TBM based on antibody detection in the CSF.

RESULTS

Thirty-six articles (58 studies) were identified. The sensitivity of antibody detection was 0.75 (95% confidence interval (CI) 0.66-0.82), specificity was 0.98 (95% CI 0.96-0.99), and the area under the receiver operating characteristic curve (AUROC) was 0.97 (95% CI 0.95-0.98). By subgroup analysis, the detection of anti-M37Ra was the highest (AUROC 0.99, 95% CI 0.98-1.00), followed by anti-antigen 5 (AUROC 0.99, 95% CI 0.97-0.99) and anti-M37Rv (AUROC 0.97, 95% CI 0.95-0.98).

CONCLUSIONS

For the early diagnosis of TBM based on antibodies in the CSF, the detection of anti-M37Ra, anti-antigen 5, or anti-M37Rv provides the greatest sensitivity and specificity.

Antibodies to antigen A60 in cerebrospinal fluid from patients with tuberculous meningitis

Cerebrospinal fluid (CSF) anti-mycobacterial antigen 60 (A60) IgM, IgG and IgA in patients affected by meningitis of different etiologies were assayed as a rapid diagnostic test in cases of tuberculous meningitis. A commercial EIA was used to test 127 CSF samples classified as follows: tuberculous meningitis (n = 27 CSF samples from 16 patients, 6 of them with AIDS), pyogenic meningitis (n = 13), non-tuberculous aseptic meningitis (n = 43) and 44 normal CSF samples (16 of them from HIV-positive patients, 8 of whom had extraneurological tuberculosis). Anti-A60 IgM was positive only in two cases (1 tuberculous meningitis and 1 self-resolving aseptic meningitis). Positive CSF anti-A60 IgG and IgA were observed in eight and nine out of 16 patients with tuberculous meningitis, but only in four and five out of 13 samples studied prior to or in the first ten days of treatment, respectively. Most of the patients with false-positive IgG and IgA (16%) had pyogenic meningitis, but without intrathecal synthesis of antibodies. In patients with aseptic meningitis, the finding of CSF anti-A60 IgG plus IgA, initially or during follow-up, can be used as a diagnostic criterion for tuberculous meningitis, with a specificity of 100%, a positive predictive value of 1, and a negative predictive value of 0.81. However, its sensitivity is only 50% in immunocompetent patients and 16% in patients with AIDS.

Improved detection of mycobacterial antigens in clinical specimens by combined enzyme-linked immunosorbent assay

Three types of antibodies against cellular and secretory-excretory protein antigens were simultaneously used for the direct detection of mycobacterial antigens in sputum and cerebrospinal fluid (CSF) specimens, using enzyme-linked immunosorbent assay (ELISA). The antibodies consisted of in-house raised and prepared anti-whole-cell, heat-killed, and sonicated Mycobacterium tuberculosis, anti-secretory-excretory protein extract of bacilli Calmette-Guerin (BCG) strain, and commercially available anti-BCG. Sputum specimens comprised 24 smear positive, culture positive, and 47 smear-negative, culture positive (SNCP), from patients with pulmonary tuberculosis, as well as 45 smear-negative, culture-negative (SNCN) control samples. The CSF specimens included 18 SNCPs from patients with tuberculous meningitis and 18 SNCN controls. The sensitivity of the individual tests for sputum and CSF specimens ranged from 70% to 79% and 72% to 89%, respectively, whereas in the combined tests it reached 86%-96% for sputum specimens and 100% for CSF specimens. The specificity of ELISAs for sputum specimens was lower in the combined (73%-87%) than in the individual (87%-98%) tests, whereas for CSF specimens it was 100% in all tests. Thus, the combined ELISA approach for mycobacterial antigen detection provides a rapid and reliable laboratory adjunct in the diagnosis of patients with tuberculosis.