Neurol India Home 

Year : 2003  |  Volume : 51  |  Issue : 4  |  Page : 474--478

Role of in vivo proton MR spectroscopy in the evaluation of adult brain lesions: Our preliminary experience

A Kumar, S Kaushik, RP Tripathi, P Kaur, S Khushu 
 Institute of Nuclear Medicine & Allied Sciences, Brig S. K. Mazumdar Road, Delhi - 110054, India

Correspondence Address:
S Kaushik
NMR Research Centre, Institute of Nuclear Medicine and Allied Sciences, Brig S. K. Mazumdar Road, Delhi - 110054,


Context: A definite diagnosis and characterization of intracranial mass lesions, based on structural Magnetic Resonance Imaging (MRI) alone may be difficult. In such cases Proton Magnetic Resonance Spectroscopy (1H-MRS) along with other non-invasive techniques represents an advance in the specificity of brain lesion diagnosis.1 Aims: The primary aim of this study was to evaluate the extent of the utility of 1H-MRS in adult brain tumors and their differentiation from similar-appearing space-occupying lesions. Material and Methods: MRS studies were performed on 1.5 Tesla whole body MR system using standard imaging head coil. Sixty patients (aged 30-65 years), including 35 males (31-65 years) and 25 females (30-65 years) were studied, along with 25 age-matched healthy volunteers (30-64 years). The Student ‘t’ test was used to statistically analyze the spectroscopic data for significant difference in the metabolite ratios of the lesions from normal brain tissue. Results: The Cho/Cr ratio was significantly raised in low and high-grade glioma and meningioma patients (1.85±0.36, 3.50±1.00 and 6.65±2.83 respectively) (mean±standard deviation), as compared with the control group (1.16±0.18); and NAA/Cr and NAA/Cho ratios were found to be lower than normal values in our study (P<0.01). However, in the non-neoplastic lesions, the Cho/Cr ratios were not statistically significant. The tubercular lesions revealed an average Cho/Cr ratio of 1.24±0.18, while it was 1.14±0.07 for infarcts. Conclusions: MR Spectroscopy was useful to arrive at a more definitive diagnosis in doubtful intracranial space-occupying lesions with similar morphological imaging patterns.

How to cite this article:
Kumar A, Kaushik S, Tripathi R P, Kaur P, Khushu S. Role of in vivo proton MR spectroscopy in the evaluation of adult brain lesions: Our preliminary experience.Neurol India 2003;51:474-478

How to cite this URL:
Kumar A, Kaushik S, Tripathi R P, Kaur P, Khushu S. Role of in vivo proton MR spectroscopy in the evaluation of adult brain lesions: Our preliminary experience. Neurol India [serial online] 2003 [cited 2020 Jul 14 ];51:474-478
Available from:

Full Text


The morphological characterization of intracranial mass lesions using conventional Magnetic Resonance Imaging (MRI) alone, even after contrast administration may be difficult without the histopathological examination of the suspected tissue.[2] Hence, there is a definite need for other non-invasive techniques to overcome this shortcoming and provide more diagnostic specificity.[1] Proton Magnetic Resonance Spectroscopy (1H-MRS) is one such technique that is superior to MRI in the detection of tumor growth in morphologically normal tissue and in the differential diagnosis of untreated intracranial space-occupying lesions (SOLs).[3]

MRS provides a detailed bio-chemical analysis (metabolites) of the tissue,[4],[5] allowing direct insight into in-vivo human brain metabolism.[5] The metabolites, reliably mapped using 1H-MRS include Choline {(Cho, 3.20 parts per million (ppm)}, Creatine (Cr, 3.02 ppm), N-acetyl-L-aspartate (NAA, 2.02 ppm), Lactate (Lac, 1.33 ppm) and Lipids (1.28-1.33 ppm).[1],[6],[7]

The primary aim of the study was to evaluate the utility of in-vivo 1H-MRS in brain tumors and their differentiation from similar appearing SOLs. Although it has been reported that no characteristic change in any particular metabolite in 1H-MRS can reliably distinguish between different tumor types,[1] an attempt has been made to differentiate between low and high-grade gliomas based on the metabolite ratios.

Material and Methods

The patients selected for the study were referred from the neurology/neurosurgery departments of various hospitals. The clinical presentations varied from features of raised intracranial tension to seizures and focal neurological signs. Patients with known extracranial malignancies and brain metastases were excluded. Sixty patients (n=60, aged 30-65 years) including 35 males [age group 31-65 years, 46.2±9.9; 46 years {mean±standard deviation (SD); median age}] and 25 females (30-65 years, 46.4±10.8; 45 years) were studied. In 20 patients, MRS was also performed from the corresponding contralateral brain regions. Twenty-five age-matched healthy volunteers (controls, aged 30-64 years) including 14 males (32-64 years, 49.1±9.7; 47 years) and 11 females (30-64 years, 45.4±11.9; 44 years) were studied for comparison and standardization. The controls selected had no apparent brain disorder or focal neurological symptomatology.

All the patients were subsequently followed up for confirmation of diagnosis. Based on histopathological findings (stereotactic/postoperative), 40 patients were diagnosed with brain tumors: Glioblastoma multiforme (GBM) (n=6); astrocytoma grade-III (n=8); astrocytoma grade-II (n=15); and meningioma (n=11). Of the remaining 20 patients, 12 had a tubercular etiology and 8 had infarcts, as evaluated on follow-up studies.

Proton MR Spectroscopy

The study was performed on 1.5 Tesla whole body MR system (Magnetom 'Vision', Siemens Erlangen, Germany) with multinuclear spectroscopic capabilities using a circularly polarized phased array head coil. Initially, each patient was subjected to routine spin echo (SE) sequences. The volume of interest (VOI) from the lesion was selected on SE-T2-weighted images for single voxel (SVS) (voxel-3.4cm[3]) and multivoxel spectroscopy {(Chemical Shift Imaging, CSI), (1-8cm[3])}. Corresponding contralateral areas and controls were also analyzed for the metabolite ratios. SVS studies were performed with Point Resolved Spectroscopy (PRESS) sequence {TR/TE/Ac (repetition time/time to echo/acquisitions) (1500/135&270/128)}; and Stimulated Echo Acquisition Mode (STEAM) sequence (1500/10/128). CSI was performed using (1500/135/2) parameters. The SVS-SE-135 spectra were used for metabolite ratio calculations.

Prior to spectroscopic measurements, global shimming to adjust for static magnetic field inhomogeneity, and local shimming {measured as Full Width at Half Maximum (FWHM)} for static and dynamic magnetic field inhomogeneities were carried out. The global shimming was optimized at 15-17 Hz, and FWHM between 5-7 Hz. Water suppression was carried out using a gaussian pulse. As far as possible, areas of edema and adjoining calvarium were avoided to prevent signal contamination. Optimal FWHM and water suppression were achieved in most cases, however, the location and heterogeneous nature of the lesions prevented optimal shimming in some patients. LUISE software (Siemens) was used for post-processing of the acquired data.

The area under the curve of a metabolite was considered as relative concentration and was measured in terms of ratios. Measuring metabolite peak area ratios has the advantage of canceling out the effects of general reduction in measured metabolite concentrations due to variations in cellular density.[8]

Statistical Analysis

The MR spectroscopic data in our study was assumed to follow normal distribution. The level of significance was determined using the Student 't' test[9] and probability value (P) 99% (P99% (P<0.01). We achieved a sensitivity of 73% and 86%, and a specificity of 94% and 96% for low and high-grade gliomas respectively.

1H-MRS plays a useful ancillary role in differentiating tumors from acute infarctions (without an inflammatory component) with an almost identical morphological pattern. In our group, infarctions were associated with perifocal edema and mild mass effect. However, MRS revealed normal Cho/Cr ratios, excluding a mitotic pathology. Reduced NAA may be observed as early as 4 days after infarction.[11] Lactate may be elevated in the initial 24 hours after a stroke, implying tissue ischemia. Chronic infarctions show decreased NAA, Cr and Cho, with no evidence of lactate or lipids.[11]

MRS is useful in the differential diagnosis of glioma vis-à-vis untreated infective brain pathologies,[3] especially in the non-inflammatory stage. Most tubercular lesions have non-specific appearances on MR imaging and may mimic other intracranial SOLs. In our study, patients with neurotuberculosis simulated tumors on imaging morphology, whereas MRS suggested a tubercular etiology. Repeat MR studies after appropriate therapy revealed significant resolution of the lesions. Tubercular lesions have been shown to exhibit strong lipid resonances, ascribed to mobile lipids within the caseous material, which are minimally visible on MR imaging.[17] In our study, Cho/Cr ratios were not significantly high, with reduced NAA/Cr and NAA/Cho ratios. Ten of these patients, however, had strong lipid peaks indicative of a possible tuberculous etiology, which was only speculative on imaging morphology.

MRS can play an important role to elucidate the cause of altered image morphology in doubtful situations, with a relatively high specificity. It also helps, at times, in distinguishing between the various types of inflammatory granulomas, which holds promise but needs further work to be critically useful. The spectral patterns are different in the majority of neoplastic and non-neoplastic intracranial mass lesions, thus aiding in better tissue characterization. Automated lesional pattern recognition using MRS is likely to be the future method of choice for classification of tumors, but these methods are not widely available as yet.[8] MRS is a useful clinical tool and more research needs to be done on this modality to let its role be defined further in terms of its clinical relevance.


1Preul MC, Caramanos Z, Collins DL, Villemure J-G, Leblanc R, Olivier A, et al. Accurate, non invasive diagnosis of human brain tumors by using proton magnetic resonance spectroscopy. Nature Med 1996;2:323-5.
2Tzika AA, Zarifi MK, Goumnerova L, Astrakas LG, Zurakowski D, Young-Poussaint T, et al. Neuroimaging in Pediatric Brain Tumors: Gd-DTPA-enhanced, Hemodynamic, and Diffusion MR Imaging Compared with MR Spectroscopic Imaging. Am J Neuroradiol 2002;23:322-33.
3Burtscher IM, Holtas S. Proton MR Spectroscopy in Clinical Routine. J Magn Res Imaging 2001;13:560-7.
4Luyten PR, Marien AJH, Heindel W, Gerwen PHJ van, Herholz K, Hollander JA den, et al. Metabolic Imaging of patients with intracranial tumors: H-1 MR spectroscopic imaging and PET. Radiology 1990;176:791-9.
5Bruhn H, Frahm J, Gyngell ML, Merboldt KD, Hanicke W, Sauter R, et al. Non invasive Differentiation of Tumors with use of Localized H-1 MR Spectroscopy in Vivo: Initial Experience in Patients with Cerebral Tumors. Radiology 1989;172:541-8.
6Poptani H, Gupta RK, Jain VK, Roy R, Pandey R. Cystic intracranial mass lesions: Possible role of in vivo MR Spectroscopy in its differential diagnosis. Magn Res Imaging 1995;13:1019-29.
7Kugel H, Heindel W, Ernestus R-I, Bunke J, Mesnil R du, Friedmann G. Human brain tumors: Spectral patterns detected with localized H-1 MR spectroscopy. Radiology 1992;183:701-9.
8Howe FA, Barton SJ, Cudlip SA, Stubbs M, Saunders DE, Murphy M, et al. Metabolic profiles of Human Brain Tumors Using Quantitative In Vivo 1H Magnetic Resonance Spectroscopy. Magn Res Med 2003;49:223-32.
9Bajaj CP, Gupta PN, editors. Elements of Statistics. 6th edn. New Delhi: R. Chand & Company; 1996. pp. 307-10.
10Burtscher IM, Skagerberg G, Geijer B, Englund E, stahlberg F, Holtas S. Proton MR spectroscopy and preoperative diagnostic accuracy: an evaluation of intracranial mass lesions characterized by stereotactic biopsy findings. Am J Neuroradiol 2000;21:84-93.
11Castillo M, Kwock L, Mukherji SK. Clinical Applications of Proton MR Spectroscopy. Am J Neuroradiol 1996;17:1-15.
12Go KG, Kamman RL, Mooyaart EL, Heesters MAAM, Pruim J, Vaalburg W, et al. Localised Proton spectroscopy and spectroscopic imaging in cerebral gliomas, with comparison to positron emission tomography. Neuroradiol 1995;37:198-206.
13Nelson SJ, Huhn S, Vigneron DB, Day MR, Wald LL, Prados M, et al. Volume MRI and MRSI Techniques for the Quantitation of Treatment Response in Brain Tumors: Presentation of a Detailed Case Study. J Magn Res Imaging 1997;7:1146-52.
14Gill SS, Thomas DGT, Bruggen NV, Gadian DG, Peden CJ, Bell JD, et al. Proton MR spectroscopy of Intracranial Tumors: In Vivo and In Vitro Studies. J Comput Assist Tomogr 1990;14:497-504.
15Kimura T, Sako K, Gotoh T, Tanaka K, Tanaka T. In: vivo single-voxel proton MR spectroscopy in brain lesions with ring-like enhancement. NMR Biomed. 2001;14:339-49.
16McKnight TR, Noworolski SM, Vigneron DB, Nelson SJ. An Automated Technique for the Quantitative Assessment of 3D-MRSI Data from Patients with Glioma. J Magn Res Imaging 2001;13:167-77.
17Gupta RK, Pandey R, Khan EM, Mittal P, Gurjral RB, Chhabra DK. Intracranial Tuberculomas: MRI signal intensity correlation with histopathology and localised proton spectroscopy. Magn Res Imaging 1993;11:443-9.