Multiparametric magnetic resonance imaging features of giant intracranial tuberculomas.

OBJECTIVES: To evaluate Magnetic Resonance Imaging (MRI) features of Giant Tuberculomas (GT) of the brain and deduce characteristic imaging phenotypes which may differentiate GT from higher grade glioma. METHODS: A retrospective analysis of MRI was done on Tuberculomas of size >2 cm. The diagnosis was established by histopathology or presumed from size reduction on follow-up MRI while on empirical anti-tubercular therapy (ATT). Multimodality characteristics of GT on T1/T2W, Fluid attenuation recovery (FLAIR), Diffusion-Weighted imaging (DWI), Susceptibility Weighted Imaging (SWI), Spectroscopy (MRS) and Perfusion weighted sequences were assessed. These imaging features were also evaluated in WHO Grade IV, IDH-wild type glioma (histopathologically and genetically proven) and a comparative analysis of the imaging features between GT and glioma was done. RESULTS: Thirty-two GT and 20 glioma were evaluated. Pronounced intralesional T2 hypointensity (n = 8;25%), T2 hyperintense crescent beneath the periphery (n = 25, 78.1%), T2W lamellated/whorled appearance (n = 17;53.125%), hyperintense rim on T1W MT (n = 25;78.1%), peripheral rim of diffusion restriction (n = 22; 68.75%), peripheral rim of blooming on SWI (n = 20, 62.5%), prominent lipid resonance on MR spectroscopy (n = 30; 93.75%), overshoot of the signal intensity-time curve above the base line (n = 9/10; 90%) on dynamic susceptibility contrast (DSC) perfusion, were remarkable imaging characteristics. Reduction of peripheral T1 hyperintensity, compaction of T2 hypointense core, expansion of sub-marginal T2 hyperintense rim and increased peripheral susceptibility (n = 20; 62.5%) during follow-up imaging, while on ATT, were standout features. GT could be differentiated from WHO grade IV (IDH-wild type) glioma on the basis of a significantly higher proportion of GTs showing a whorled/lamellated appearance, T1 hyperintense rim, T2 hypointense core, DWI-ADC mismatch, well-defined rim on SWI, prominent lipid peak on MRS and a submarginal T2 hyperintense rim. GT showed a higher normalized ADC ratio from the core as well as the rim. Significantly higher proportion of glioma showed a T1 hypointense and T2 hyperintense core and a nodular rim enhancement. A significantly higher r CBV, Choline to creatine, choline to NAA ratio and mean thickness of the peripheral enhancing rim were defining features among gliomas. CONCLUSION: Neuroimaging features of GT have been elucidated. Reduction of peripheral T1 hyperintensity, compaction of T2 hypointense core, expansion of sub-marginal T2 hyperintense rim, and increased peripheral susceptibility on follow-up may be considered imaging markers of response to anti-tubercular therapy. Multiparametric MRI features can differentiate GT from WHO grade IV (IDH-wild type) glioma.

Modulation of angiogenic factor VEGF by DNA-hsp65 vaccination in a murine CNS tuberculosis model.

Tuberculosis (TB) is a serious public health problem. Development of experimental models and vaccines are essential to elucidate physiopathological mechanisms and to control the disease. Vascular endothelial growth factor (VEGF) is a potent activator of vascular permeability and angiogenesis. VEGF seems to participate in breakdown of the blood brain-barrier (BBB) in tuberculous meningitis (TBM), contributing to worsening of disease. Therefore, the objective here was to extent the characterization of our previously described murine model of central nervous system TB (CNS-TB) by describing the VEGF participation in the CNS disease, and suggesting a vaccination plan in mice. Plasmid encoding DNA protein antigen DNA-hsp65 has been described as a protector against TB infection and was used here to test its effectiveness in the prevention of VEGF production and TB disease. Vaccinated mice and its controls were injected with Mycobacterium bovis bacillus Calmette-Guerin (BCG) in cerebellum. Four weeks after BCG injection, mice were perfused and brains were paraffin-embedded for VEGF expression analysis. We observed VEGF immunohistochemical expression in TBM and granulomas in non-vaccinated mice. The DNA-hsp65 treatment blocked the expression of VEGF in mice TBM. Therefore, our murine model indicated the VEGF participation in the physiopathology of CNS-TB and the potential prevention of the DNA-hsp65 in the disease progression.

Differentiation of intracranial tuberculomas and high grade gliomas using proton MR spectroscopy and diffusion MR imaging.

OBJECTIVE: The purpose of this study was to determine whether proton MR spectroscopy ((1)H MRS) and diffusion-weighted (DW) imaging can be used to differentiate intracranial tuberculomas from high grade gliomas (HGGs). MATERIALS AND METHODS: A total of 41 patients (19 with intracranial tuberculomas and 22 with HGGs) were examined in our study. (1)H MRS and DW imaging were performed at a 1.5T MR scanner before operation or treatment. Concentrations of N-acetylaspartate (NAA), creatine (Cr), choline (Cho), and lipid and lactate (LL) in the contrast-enhancing rim of each lesion were expressed as metabolite ratios and were normalized to the contralateral hemisphere. The apparent diffusion coefficient (ADC) was also calculated. The metabolite ratios and ADC values in the enhancing rim of intracranial tuberculomas and HGGs were compared using the Wilcoxon rank sum test. Diagnostic accuracy was compared using receiver operating characteristic (ROC) analysis. RESULTS: Significant differences were found in the maximum Cho/Cr (P=0.015), Cho/NAA (P=0.001) and Cho/Cho-n ratios (P=0.002), and minimum ADC value (P<0.001) between the intracranial tuberculomas and HGGs. Diagnostic accuracy was higher by minimum ADC value than maximum Cho/Cr, Cho/NAA and Cho/Cho-n ratios (93.8% versus 75.7%, 80.8% and 78.1%). CONCLUSION: These results suggest a promising role for (1)H MRS and DW imaging in the differentiation between the intracranial tuberculomas and HGGs.

Dynamic contrast-enhanced (DCE) derived transfer coefficient (ktrans) is a surrogate marker of matrix metalloproteinase 9 (MMP-9) expression in brain tuberculomas.

PURPOSE: To correlate dynamic contrast-enhanced (DCE) MRI derived perfusion indices with immunohistochemically obtained vascular endothelial growth factor (VEGF) and matrix metalloproteinase-9 (MMP-9) in a cellular fraction of brain tuberculomas (BT). MATERIALS AND METHODS: Thirteen BT patients underwent DCE MRI. Perfusion indices (cerebral blood volume [CBV], transfer coefficient [ktrans] and leakage [ve]) maps were generated for the quantitative analysis. The CBV was corrected for the leaky blood-brain barrier (BBB). The relative CBV (rCBV), ktrans, and ve were calculated by placing 10 regions of interest (ROIs) showing the highest values in the lesion. The percentage area of VEGF and percentage area of MMP-9 and microvessel density (MVD) were quantified from 10 fields per lesion with maximal expression of the excised BT. Pearson correlation analysis between physiological indices and quantitative VEGF, MMP-9, and MVD was performed for each ROI. RESULTS: The average value of rCBV, ktrans, ve, VEGF, MMP-9, and MVD were 3.53+/-0.37, 2.04+/-0.40 min(-1), 0.71+/-0.09, 12.51+/-2.56, 18.09+/-2.06, 10.87+/-1.99, respectively. The ktrans (r=0.918, P<0.001) and ve (r=0.899, P<0.001) showed significant correlation with MMP-9, while rCBV correlated significantly with MVD (r=0.962, P<0.001) and VEGF (r=0.868, P<0.001). CONCLUSION: We conclude that the expression of MMP-9, a marker of BBB disruption and disease activity in BT correlates with DCE-derived ktrans and thus has the potential to be used as its surrogate marker.

NMR spectroscopic identification of cholesterol esters, plasmalogen and phenolic glycolipids as fingerprint markers of human intracranial tuberculomas.

Detailed (1)H and (13)C NMR spectroscopy of lipid extracts from 12 human intracranial tuberculomas and two control brain tissue samples was performed to assess the role of lipids in the disease process. One-dimensional and two-dimensional NMR techniques were used to resolve the mixture of lipid components and make resonance assignments. The lipid components that could be identified in tuberculoma lipid extracts and not in control samples were: cholesterol ester, plasmalogen and phenolic glycolipids. It is proposed that the combined occurrence of these lipid components can be used as 'fingerprint markers' for the differentiation of intracranial tuberculoma from healthy brain tissue. Furthermore, phenolic glycolipids present in intracranial tuberculomas may have diagnostic significance in differentiating them from other disease conditions of the central nervous system such as malignant tumors.

Inflammatory granulomas: evaluation with proton MRS.

Intracranial inflammatory lesions consisting mainly of tuberculomas (n = 28), neurocysticercosis (NCC) (n = 10), and non-specific inflammatory granulomas (IG) (n = 22) were evaluated with proton MRS. Water-suppressed proton spectra were acquired from 60 patients using the STEAM sequence with an echo time of 135 ms and metabolite ratios determined from the spectra. Student's paired t-test and chi2-test were used to analyse the data. Statistically significant differences were observed for the following ratios between the three patient categories: NAA/ Cr (p < 0.0001), NAA/Cho (p = 0.001) and Cr/Cho (p = 0.02) for the non-specific IG and NCC, NAA/Cho (p = 0.03) for non-specific IG and tuberculoma, and NAA/Cr (p < 0.0001) for NCC and tuberculoma. While lipids were seen in 86% of the tuberculomas, they were observed in only 20% of the NCCs (chi2 = 6.81, p = 0.009), and 21% of the non-specific IGs (chi2 = 10.75, p = 0.0001). While the presence of lipid can be used for differentiating tuberculomas from both non-specific IG and NCC, the extremely low levels of metabolites together with a poor signal/noise ratio could itself act as a marker for NCC.

Diagnostic assessment of brain tumours and non-neoplastic brain disorders in vivo using proton nuclear magnetic resonance spectroscopy and artificial neural networks.

PURPOSE: Experiments were carried out to assess the potential of artificial neural network (ANN) analysis in the differential diagnosis of brain tumours (low- and high-grade gliomas) from non-neoplastic focal brain lesions (tuberculomas and abscesses), using proton magnetic resonance spectroscopy (1H MRS) as input data. METHODS: Single-voxel stimulated echo acquisition mode (STEAM) (echo time of 20 ms) spectra were acquired from 138 subjects including 15 with low-grade gliomas, 47 with high-grade gliomas, 18 with tuberculomas, 18 with abscesses and 40 healthy controls. Two neural networks were constructed using the spectral points from 0.6 to 3.4 parts per million. In the first network construction, the ANN had to differentiate between tumours from infections, while the second network had to differentiate between all five histological classes. RESULTS: ANN analysis gave a histologically correct diagnosis for low- and high-grade gliomas with an accuracy of 73% and 98% respectively. None of the 62 tumours was diagnosed as an infectious lesion. Among the non-neoplastic lesions, ANN classification was correct in 89% of tuberculomas and in 83% of brain abscesses. The specificity of ANN diagnosis was 98%, 92%, 99%, and 100% for low-grade gliomas, high-grade gliomas, tuberculomas and abscesses respectively. CONCLUSION: The present data show the clinical utility of non-invasive 1H MRS by automated ANN analysis in the diagnosis of tumour and non-tumour cerebral disorders.