Evaluation of MRI based diagnostic algorithm for intra-axial brain masses: A single centre experience

Abstract

Background and Purpose: Preoperative evaluation and differentiation of intra-axial brain masses is essential for therapy and prognosis. The integrated Magnetic Resonance Imaging (MRI) techniques such as Diffusion Weighted Imaging (DWI), Magnetic Resonance Spectroscopy (MRS) and Dynamic Susceptibility Contrast (DSC) MRI allow insights into tumour anatomy and physiology. Our study was designed to find the sensitivity, specific and accuracy of a diagnostic algorithm comprising of these techniques. Materials and Methods: Thirty consecutive immunocompetent patients with treatment naive intra-axial brain masses underwent MRI from July 2016 to July 2017. Conventional and DWI, MRS and MR perfusion scans were performed using 1.5 T MR imaging system. The following cut-off values (ADC ?1.1 /100mm2, intralesional Cho/NAA > 2.2, perilesional Cho/NAA >1 and rCBV >1.75) were applied to the diagnostic strategy algorithm. Histopathological analysis served as the gold standard. Sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV) and accuracy were obtained. Results: Twenty one of 30 cases were correctly classified using the diagnostic algorithm. One case of primary central nervous system lymphoma (PCNSL) was misdiagnosed as metastasis, 2 cases of high-grade neoplasm (HGN) as brain abscesses, 1 case of HGN as metastasis, 1 case of HGN as low-grade neoplasm (LGN), 1 case of LGN as HGN, 1 case of LGN as abscess / tumefactive demyelinating lesion (TDL), 1 case of brain abscess as HGN, and 1 case of metastasis (breast carcinoma) as HGN. Conclusion: Diagnosing intra-axial brain masses using a combination of advanced imaging techniques (MRS, MR perfusion and ADC) is fairly accurate.