Although only 1% of TB cases involve the CNS, these cases represent the most severe form of the disease, leads to serious permanent neurological damage, even lethal complications,11-13particularly in HIV-positive patients.Therefore, efficient diagnostic strategy of BT is imperative in minimizing the risk of complications and improving overall outcomes by virtue of timely diagnosis and treatment.
In terms of clinical features, similar to other occupied lesions in the brain, BT might exert local mass effect on brain tissue, leading to headache, vomiting, decreased consciousness, focal neurological signs and seizures.14 So, there is no specificity of the clinical features in BT.
The diagnosis of BT is often with the help of imaging techniques such as CT and MRI.4 Brain imaging of BT frequently manifest clinically silent single or multiple CNS granulomata with or without meningitis. On CT, the characteristic appearance is a nodular enhancing lesion with a central hypodense region; on MRI, the early focal cerebritis stage is marked by edema and ill-defined enhancement, the later mature stage by central hypointensity and peripheral enhancement.3,15,16 However, several diseases are also capable of producing similar imaging features, such as cysticercosis, toxoplasmosis, fungal granuloma, bacterial abscesses, brain neoplasms (gliomas, lymphomas and brain metastases).16 In our study, merely according to imaging features, three patients (case 10, 13, 15) treated with anti-toxoplasma, two patients (case 3, 12) treated as cerebral infarction prior to hospitalization. Thus, relying solely on neuroimaging, which is difficult to differentiate BT from other brain parenchymal lesions, even harmful.
IGRAs is among the commonly used M.TB tests available. Interferon-gamma (IFN-γ) is released by CD4+ T-cells, which induce macrophage activation and promote destruction of mycobacteria.17 Advantages of IGRAs include latent tuberculosis infection detection and specificity that is relatively unaffected by the TB vaccine. Simmons et al. reported IGRAs-positive rate of 50% among culture-confirmed cases of TBM.18 Nevertheless, these assays do not differentiate between latent and active M.TB infection,19 also the diagnostic sensitivity of IGRAs will be reduced in immunocompromised persons. Hence, in this mainly HIV-positive patients’ study, IGRAs’ sensitivity is 35.7% (5/14).
Culture identification for M.TB is always the gold standard for diagnosing BT, moreover, which allows for drug sensitivity testing, essential for appropriate treatment. However, it has low sensitivity,15 which is affected by various factors such as paucibacillary as nature of disease,20 type of species, experience of the technical person, and so on.5 Besides that, M.TB is particularly slow growing with the time for about 4-8 weeks.21 The indolence hinders MTB culture detection within a clinically relevant time frame, since immediate treatment is vital for BT’s prognosis. In our study, only 6 (42.9%, 6/14) TTSs for culture positive, besides the reasons above mentioned, 6 of the 8 culture negative patients were HIV positive, which may be due to the statement that bacillary load of M.TB is usually low in HIV-positive patients,22 that compromising sensitivity of bacteriological tests.
PCR is an effective method for the rapid detection of specific bacterial DNA in clinical specimens, which can be useful to improve the speed and accuracy in detecting M.TB. However, the reliability of PCR testing for M.TB DNA is not well established, primarily because of variability in sensitivity and specificity across multiple laboratories. A meta-analysis of PCR of CSF for TBM showed a pooled sensitivity of 56% and specificity of 98%.6 Another article has proven high specificity (80-100%), a wider sensitivity range (30-100%) has been shown.23 The low sensitivity of PCR is due to the PCR inhibitors such as host proteins, blood and even eukaryotic DNA in extrapulmonary specimens.24 The sensitivity of PCR of TTSs was 62.5% (10/16) in this study, the 6 negative TTSs were also negative by culture, which may also attribute to low bacillary load of M.TB in HIV-positive patients22 (4 of the 6 negative patients were HIV positive).
Gene Xpert is a rapid automated molecular test with high accuracy, reduces hands-on time, decreased risk of cross contamination and the ability to identify rifampicin resistance for pulmonary25 and various extrapulmonary samples of TB such as CSF, urine, lymph node and other tissues,7,26 which is recommended by World Health Organization (WHO) for TB detection.In a meta-analysis of 30 studies indicated Xpert MTB/RIF pooled sensitivity and specificity against culture were 71.1% (62.8 to 79.1) and 96.9% (95.4 to 98.0)7. Another meta-analysis of 13 studies in which Gene Xpert was evaluated against culture, the pooled sensitivity was 80.5% and specificity 97.8%.27 In our series, 8 positives in 10 Gene Xpert of TTSs, with the sensitivity of 80.0%. Nevertheless, despite Gene Xpert is more sensitive than other co-existing conventional methods,28 still can’t absolute to identify all cases of CNS-TB.
There are few studies reported on GeneXpert or PCR for the detection of M.TB in TTS, especially in HIV-positive patients heretofore. However, when BT is the only lesion, didn’t complicated by meningitis, definite diagnosis of BT using CSF remains difficult, as M.TB may be encapsulated by fibrous of the caseous granulomas.3,16 A study performed by Dil-Afroze et al. showed CSF-PCR was positive for 21/27 patients (77.7% sensitivity) of TBM patients, only 6/20 patients (30% sensitivity) with BT were positive.29 In our study, of the 11 patients available for CSF-PCR, only 1 patient was positive. This limitation of diagnostic strategy led to missed diagnosis of BT.30
Histopathological analysis combined with culture identification, remains the gold standard for diagnosis of BT. Tuberculomas are granulomatous mass lesions composed of a central zone of caseation surrounded by a collagenous tissue capsule.31 Whereas, histopathology is technically demanding and time consuming,8 the sensitivity reported ranges from 50% to 60%.8,32 There are times when the pathology simply indicates that chronic inflammatory granuloma, which is insufficient for the diagnosis of tuberculosis, because this pathological feature may also be found in other infectious and non-infectious disease. In our series, there were 7 patients only presented with non-specific glial cells hyperplasia and infiltration with inflammatory cells. In addition, histopathology does not distinguish between BT and other granulomatous diseases such as sarcoidosis, leprosy and systemic lupus erythematosus.33 Research from India, including only histopathologically verified cases of BT, revealed a lower rate of 54% complete resolution by 24 months of ATT,34 which mainly attribute to the failure of pathological verification.
The sensitivity of AFB stain in various published studies was reported in the range of 20% to 60%,5 which can be increased to a small extent using special physical and/or chemical methods of sample processing and special equipment like fluorescent microscopy.35 More than that, at times, we also need to distinguish between M.TB and mycobacterium leprae, which was also positive by AFB stain. The AFB stain of TTSs’ sensitivity in our study was 68.8% (11/16).
Since these test methods above can’t offer sufficient sensitivity to confidently detect M.TB alone5, however, combined-diagnostic methods of TTS, may be considered superior in diagnosing of BT.11