Sensitivity and Specificity of Magnetic Resonance Spectroscopy (MRS) in the Diagnosis of Meningiomas

Abstract

Background: Meningiomas are the most common extra axial tumors of brain accounting 13-26% of primary intracranial neoplasm. They are the most common non glial tumor of central nervous system (CNS). Meningiomas arise from the arachnoid cells called meningothelial cell and appear as extra axial, sharply demarcated, solid mushroom imaging pattern with dense homogenous contrast enhancement on CT and MRI. Objective: To determine the sensitivity and specificity of Magnetic Resonance Spectroscopy (MRS) in the Diagnosis of Meningiomas. Methods and Materials: This cross-sectional study was conducted on 45 patients with suspected meningioma who were referred to the Radiology and Imaging Department of Bangabandhu Sheikh Mujib Medical University (BSMMU) from the Department of Neurology and Neurosurgery during period from October 2019 to September 2021. Magnetic Resonance Spectroscopy (MRS) findings were evaluated taking into account the histopathology as a gold standard. Statistical analyses were carried out by using SPSS version 23.0. The quantitative observations were indicated by frequencies. Receiver operating characteristic (ROC) curve was generated to determine the cutoff value for the best sensitivity, specificity, negative and positive predictive values of choline /N-acetyl aspartate ratio with regard to diagnosis and characterization of meningiomas. Results: In MRS out of 45 patients 36(80%) patients were diagnosed as meningioma among them 30(67%) patients had increased alanine, 38(84%) patients had increased choline, 22(49%) patients NAA had not detected, 20(44%) patients NAA were minimally detected. Remaining 9(20%) patients had non-meningioma. In histopathology out of 45 patients 38(84%) patients had meningioma and 7(16%) patients had non meningioma. Histopathology showed 27(71%) patients was grade I, 3(8%) was grade II and 8(21%) was grade III. Increased lactate was detected in 11 patients, out of which 9(82%) patients were found high grade and 2(18%) patients were low grade meningiomas. The validity of meningioma by MRS was represented by calculating sensitivity, specificity, accuracy, positive and negative predictive values taking into account histopathology as gold standard. Sensitivity was 89.47%, specificity was 71.43%, accuracy was 86.67%, PPV was 94.44% and NPV was 55.56%. Based on the receiver-operator characteristic (ROC) curve choline/N-acetyl aspartate ratio had area under curve 0.850. Receiver-operator characteristic (ROC) was constructed by using choline /N-acetyl aspartate ratio, which gave a cut off value ≥2.99, with 80.0% sensitivity and 62.5% specificity for prediction of high grade. Conclusion: The study suggested that MRS findings were consistent with low specificity in the diagnosis of meningiomas but MRS can characterize the types of meningiomas keeping histopathology as a gold standard. Increased alanine, decreased or absent NAA, increased choline, decreased creatine could be the spectroscopic parameters for the diagnosis of meningiomas and by using choline/N-acetyl aspartate ratio more sensitivity was found to predicted high grade along with lactate peak.