A prospective study to evaluate the role of multiparametric magnetic resonance imaging in the grading of gliomas using magnetic resonance imaging perfusion and diffusion and multivoxel magnetic resonance spectroscopy

Abstract

Background: The primary aim of this study was to evaluate the individual and combined efficacy of magnetic resonance imaging (MRI) parameters, which include MRI perfusion, MRI diffusion-weighted imaging (DWI) and magnetic resonance spectroscopy (MRS) in the grading of gliomas as low grade versus high grade. The pre-operative imaging-based grading of gliomas by multiparametric MRI was compared with the gold standard histopathological studies. Methods: A total of 22 patients referred to the radiology department for multiparametric MRI of the brain with presumptive diagnosis of glioma on computed tomography/MRI were included in the study. Conventional T1, T2 and fluid-attenuated inversion recovery images were obtained followed by perfusion MRI using gadopentetate dimeglumine (Magnevist) administration. This was followed by DWI and MRS. Results: Our statistical analysis demonstrated that a cut-off of apparent diffusion coefficient value of 954.085 (10–6 mm2/Sec) provides a sensitivity and specificity of 87.5% and 85.7%, respectively, in differentiating low-grade gliomas (LGGs) from high-grade gliomas (HGGs). A choline/creatine ratio cut-off value of 2 provides sensitivity and specificity of 100% and 92.9%, respectively, while a cut-off value of 1.45 of choline/N-acetylaspartate ratio provides both sensitivity and specificity of 100% in differentiating LGG from HGG. A cut-off of 1.9 for maximum relative cerebral blood volume (rCBV) value provides both sensitivity and specificity of 100% in differentiating LGGs from HGGs. Conclusions: We concluded that perfusion MRI (rCBV) was the best parameter among perfusion MRI, DWI and MRI spectroscopy in differentiating HGGs from LGGs. Combined multiparametric results showed a diagnostic accuracy, sensitivity, specificity, positive predictive value and negative predictive value of 86.4%, 82.4%, 100%, 100% and 62.5%, respectively, on comparison with gold standard histopathological grading.