Author/Authors :
Matsusue, Eiji Department of Radiology - Tottori Prefectural Central Hospital, Japan , Inoue, Chie Department of Radiology - Tottori Prefectural Central Hospital, Japan , Tabuchi, Sadaharu Department of Neurosurgery - Tottori Prefectural Central Hospital, Japan , Yoshioka, Hiroki Department of Neurosurgery - Tottori Prefectural Central Hospital, Japan , Nagao, Yuichiro Department of Neurosurgery - Tottori Prefectural Central Hospital, Japan , Matsumoto, Kensuke Department of Radiology - Tottori Prefectural Central Hospital, Japan , Nakamura, Kazuhiko Department of Radiology - Tottori Prefectural Central Hospital, Japan , Fujii, Shinya Division of Radiology - Department of Multidisciplinary Internal Medicine - Tottori University, Japan
Abstract :
Background
Proton magnetic resonance spectroscopy (MRS) provides structural and metabolic information that is useful for the diagnosis of meningiomas with atypical radiological appearance. However, the metabolite that should be prioritized for the diagnosis of meningiomas has not been established.
Purpose
To evaluate the differences between the metabolic peaks of meningiomas and other intracranial enhanced mass lesions (non-meningiomas) using MR spectroscopy in short echo time (TE) spectra and the most useful metabolic peak for discriminating between the groups.
Material and Methods
The study involved 9 meningiomas, 22 non-meningiomas, intracranial enhancing tumors and abscesses, and 15 normal controls. The ranking of the peak at 3.8 ppm, peak at 3.8 ppm/Creatine (Cr), β-γ Glutamine-Glutamate (bgGlx)/Cr, N-acetyl compounds (NACs)/Cr, choline (Cho)/Cr, lipid and/or lactate (Lip-Lac) at 1.3 ppm/Cr, and the presence of alanine (Ala) were derived. The metabolic peaks were compared using the Mann-Whitney U test. ROC analysis was used to determine the cut-off values for differentiating meningiomas from non-meningiomas using statistically significant metabolic peaks.
Results
The ranking of the peak at 3.8 ppm among all the peaks, peak at 3.8 ppm/Cr, bgGlx/Cr, Lip-Lac/Cr, and the presence of Ala discriminated meningiomas from non-meningiomas with moderate to high accuracy. The highest accuracy was 96.9% at a threshold value of 3 for the rank of the peak at 3.8 ppm.
Conclusion
A distinct elevated peak at 3.8 ppm, ranked among the top three highest peaks, allowed the detection of meningiomas.
Keywords :
Magnetic resonance imaging , magnetic resonance spectroscopy , meningioma , 3.8 ppm , short TE