Comparative Evaluation of Brain Tuberculosis and Metastases Using Combined Analysis of Arterial Spin Labeling Perfusion and Diffusion Tensor Imaging

Abstract

Purpose: To differentiate intra-axial tuberculomas (TB) from metastases based on quantitative differences in the perfusion and diffusion indices of lesion and perilesional edema using arterial spin labeling (ASL) and diffusion tensor imaging (DTI) techniques. Materials and methods: This prospective study included newly diagnosed untreated 12 patients of TB and 13 of metastases who underwent routine magnetic resonance imaging including DTI and ASL sequences. A region of interest analysis was performed and cerebral blood flow (CBF) values of lesion (L), perilesional edema (PE), and normal contralateral white matter (CWM) were calculated. To account for individual patient variation CBF values were normalized (n) to CWM to obtain the nCBFL and nCBFPE ratios. Similarly, DTI data was processed to obtain fractional anisotropy (FA), mean diffusivity, radial diffusivity, and axial diffusivity values from the lesion and PE. Results: Metastatic lesions revealed statistically significant (p = 0.001) high values of median nCBFL than TB whereas the difference in the median nCBFPE was not statistically significant (p = 0.174). TB showed higher median FAL compared to metastases (p = 0.031) while no statistically significant difference was found in mean values of other diffusion parameters such as mean diffusivity, radial diffusivity and axial diffusivity. Analysis by the receiver operating characteristic curve method revealed a cut-off value of ≥2.865 for nCBFL (Sensitivity = 0.85, Specificity = 0.84, positive predictive value (PPV) = 0.85, Negative predictive value (NPV) = 0.83) and ≤0.073 for FAL (Sensitivity = 0.77, Specificity = 0.58, PPV = 0.67, NPV = 0.70) in differentiating metastases from TB. Conclusion: Combined analysis of noncontrast ASL perfusion and DTI technique may markedly benefit in differentiation of TB from metastases.