|Year : 2016 | Volume
| Issue : 1 | Page : 23-32
Intracranial space occupying lesions: could differentiation be reached without biopsy?
Mahmoud A Dawoud1, Reda A Al-Arabawy1, Ali Ibrahim M Seif Eldeinb2, Noha A Darwish3
1 Department of Radiodiagnosis and Medical Imaging, Faculty of Medicine, Tanta University, Tanta, Egypt
2 Department of Neurosurgery, Faculty of Medicine, Tanta University, Tanta, Egypt
3 Department of Radiodiagnosis and Medical Imaging, Basyion General Hospital, Egypt
|Date of Web Publication||15-Apr-2016|
Reda A Al-Arabawy
Department of Radiodiagnosis and Medical Imaging, Faculty of Medicine, Tanta University, Tanta
Definite diagnosis and characterization of intracranial mass lesions based on structural MRI alone may be difficult. In such cases, proton magnetic resonance spectroscopy (MRS) represents an advance in the specificity of brain lesion diagnosis.
Aim of the work
The aim of this study was to evaluate the role of MRS in the differentiation of intracranial space occupying lesions (benign/malignant) and malignant (low-high grade)/metastasis.
Patients and methods
The study was conducted on 40 patients with intracranial space occupying lesions. All patients were subjected to MRS
The ratios were calculated, including Cho/NAA and Cho/Cr, in both intralesional and perilesional regions.
MRI and MRS of the patients were compared with histopathological analysis and we found the following results: conventional MRI had successfully diagnosed 27 cases as primary tumors and three cases as metastasis. Conventional MRI had failed to take decision in six cases as to whether they were primary or metastatic and four cases were diagnosed as non-neoplastic lesions. MRS had correctly diagnosed 31 cases as primary tumors of the 40 cases; five cases were diagnosed as metastasis and four cases were diagnosed as non-neoplastic lesions.
MRS provides information on the metabolic state of brain tissue. Thus, it is useful to arrive at a more definitive diagnosis compared with MRI in doubtful intracranial space occupying lesions with similar morphological imaging patterns.
Keywords: intracranial, space occupying lesions, magnetic resonance spectroscopy
|How to cite this article:|
Dawoud MA, Al-Arabawy RA, Seif Eldeinb AM, Darwish NA. Intracranial space occupying lesions: could differentiation be reached without biopsy?. Tanta Med J 2016;44:23-32
|How to cite this URL:|
Dawoud MA, Al-Arabawy RA, Seif Eldeinb AM, Darwish NA. Intracranial space occupying lesions: could differentiation be reached without biopsy?. Tanta Med J [serial online] 2016 [cited 2019 Dec 9];44:23-32. Available from: http://www.tdj.eg.net/text.asp?2016/44/1/23/180554
| Introduction|| |
Intracranial tumors are a significant health problem. The annual incidence of primary and secondary central nervous system neoplasms ranges from 10 to 17 per 100 000 individuals  .
A definite diagnosis and characterization of intracranial mass lesions based on structural MRI alone may be difficult. In such cases, proton magnetic resonance spectroscopy (1H-MRS) along with other noninvasive techniques represents an advance in the specificity of brain lesion diagnosis  .
1H-MRS gives completely different information related to cell membrane proliferation, neuronal damage, energy metabolism, and necrotic transformation of brain or tumor tissues  .
Thus, 1H-MRS 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)  .
MRS provides a detailed biochemical analysis (metabolites) of the tissue, allowing direct insight into in-vivo human brain metabolism. 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 (1.33 ppm), and lipids (1.28-1.33 ppm). Alanine (1.5 ppm) and acetate (1.92 ppm) were also reported  .
Non-neoplastic lesions such as brain abscesses are marked by decreases in Cho, Cr, and NAA. In patients with epidermoid cyst, lactate along with an unassigned resonance at 1.8 ppm was reported and could be easily differentiated from arachnoid cyst, which show minimal lactate  .
Only lactate is commonly observed in a variety of intracranial cystic masses, except for abscess and cysticercosis, in which resonances of acetate, succinate, aminoacids, and/or unassigned metabolites can be seen in addition to a lactate peak. Tuberculous lesions have been shown to exhibit strong lipid resonances, ascribed to mobile lipids within the caseous material, which are minimally visible on MRI  .
As for brain neoplasms, proton MR spectra obtained from them typically show the following: (a) decreased NAA, a marker of neuronal integrity; (b) diminished Cr, involved in cellular energetics and osmotic balance; and (c) increased Cho, involved in cell membrane turnover. Lactate and mobile lipids can be evident in aggressive tumors, reflecting increased anaerobic metabolism and cellular necrosis, respectively  .
Gliomas exhibit significantly increased Cho and lipid formation, with higher WHO tumor grading. In addition, the Cho/Cr ratio of 1H-MRS provides additional information to MRI in differentiating residual/recurrent gliomas from non-neoplastic lesions (radiation necrosis), being higher in the former. Metastases have elevated Cho similar to anaplastic astrocytomas, but can be differentiated from high-grade gliomas by their higher lipid levels  .
MRS provides information on the metabolic state of brain tissue. Thus, it is useful to arrive at a more definitive diagnosis in doubtful intracranial SOLs with similar morphological imaging patterns  .
Our aim was to evaluate the role of MRS in the differentiation of intracranial SOLs (benign/malignant) and malignant (low-high grade)/metastasis.
| Patients and methods|| |
The study was conducted on 40 patients with intracranial SOLs diagnosed using MRI in Tanta University Hospitals at the diagnostic radiology and medical imaging department at a period of 10 months.
Patients with intracranial SOLs were included in the study.
Patients who had contraindication to MRI were excluded from the study.
All patients were subjected to the following:
Magnetic resonance imaging
- First, patients had to provide informed consent. Any expected risk was explained to the patients. Privacy of all patients' data was guaranteed.
- Second, proper history taking was carried out, including personal, present, and past history.
- Third, general examination and recording of vital signs were carried out.
- Neurological examination was carried out for all patients.
All MRI scans were performed using a 1.5 T unit (Signa high speed; General Electric Medical Systems, USA) using a head coil at the MRI unit in diagnostic radiology and medical imaging department, Tanta University Hospitals. The slice thickness was 6 mm, the matrix was 256 × 256, and the field of view was 220-240 mm.
Magnetic resonance spectroscopy
- Axial T1WI (TR/TE = 400-600/10-20 m/s).
- Axial T2WI (TR/TE = 2000-400/100-120 m/s).
- Coronal and/or axial FLAIR images (TR/TE/Inversion time (TI) = 4000-6000/140/1200).
- Postcontrast T1WI study was conducted after injection of gadolinium-based contrast agents such as Magnevist (0.1 ml/kg).
Multivoxel MRS was performed using a spin-echo mode sequence (SE) with long TE (144 mm/s) and short TE (35 mm/s). Water suppression was achieved with the chemical shift selection technique.
The voxels were placed on the lesions, perilesional areas, and at normal region away from cerebrospinal fluid and scalp fat to avoid contamination.
The metabolites were identified, including NAA at 2.0 ppm, Cr at 3.0 ppm, Cho at 3.2 ppm, lipid at the range of 0.7-1.3 ppm, lactate at 1.33 ppm, and myo inositol at 3.56 ppm.
The ratios were calculated, including Cho/NAA and Cho/Cr, in both intralesional and perilesional regions.
Biopsy and histopathology were carried out in all neoplastic lesions.
Statistical presentation and analysis of the present study was conducted using the mean, SD, and the χ2 -test using SPSS V.20 (SPSS Inc., Chicago, Illinois, USA).
The χ2 -test is used for comparison between two groups as regards qualitative data:
where: ∑ = summation, O = observed value and E = expected value.
| Results|| |
Forty patients were included in this study with intracranial SOLs diagnosed using MRI in diagnostic radiology and medical imaging department of Tanta University Hospitals from August 2014 to May 2015.
There were 24 male (60%) and 16 female (40%) patients and their ages ranged from 8 to 66 years [Table 1].
Patients' complains in the studied cases
Clinically, among the 40 studied cases, 28 patients (70%) complained of headache, 20 patients (50%) had cerebral dysfunction (disturbed conscious level or generalized convulsions), 12 patients (30%) were suffering from hemiplegia or hemiparesis, four patients (10%) had visual complaints (diminution of vision or diplopia), and two patients (5%) had behavioral changes (χ2 = 22.820; P value = 0.001).
In this study, as regards the distribution of the intracranial SOLs, 18 patients (45%) had lesions in the parietotemporal regions, 12 patients (30%) had lesions in the occipital regions, eight patients (20%) had lesions in the frontal regions, and two patients (5%) had lesions invading and crossing mid line through the corpus callosum [Figure 2] (χ2 = 18.130; P value = 0.001).[Figure 1]
|Figure 1: Case 1: An MRI showing left temporoparietal intra-axial space occupying lesion; it exhibited low signals on T1WIs (a) and high signals on T2WIs and fl air sequences (b, c). An MRS showing intralesional voxels of long TE 144 (d); it showed increased Cho and decreased NAA and creatine, with mild lipid and lactate in short TE (e) and a Cho/NAA ratio of 3.2 and a Cho/ Cr ratio of 5. The above MRS and MRI are in favor of low-grade tumor. Histological examination revealed astrocytoma of WHO grade II|
Click here to view
|Figure 2: Case 2: MRI showing an ill-defined, irregularly shaped, intra-axial space occupying lesions in the left basal ganglia and the left and the right thalamus. It exhibited low signals on T1WIs (a) and high signals on T2WIs and flair signal (b, c). An MRS showing intralesional voxels (d) with increased Cho, with a Cho/NAA ratio of 1.7 and Cho/Cr ratio of 2. The above MRS and MRI are in favor of low-grade primary brain tumor. Histological examination revealed primary CNS lymphoma (lymphomatous infiltration on the left and right).|
Click here to view
The number of lesions in the studied cases was as follows: 35 patients (87.5%) had solitary lesion and five patients (12.5%) had multiple lesions [Figure 3] (χ2 = 25.690; P value = 0.001).
|Figure 3: Case 3: An MRI showing left temporal space occupying lesion invading the insular cortex: it exhibited low signals on T1WIs and high signals on T2WIs and fl air WIs (a, b), with faint enhancement in the postcontrast study (c). An MRS showing voxels taken inside the lesion of long TE 144 (d). It showed marked decrease in NAA and Cr and increase in Cho, with the presence of lipids and lactate and a Cho/NAA ratio of 4.8 and a Cho/Cr ratio of 3. (e) Perilesional voxel shows decreased Cr and NAA. The above MRS and MRI are in favor of Gliomatosis cerebri. Histological examination revealed Gliomatosis cerebri|
Click here to view
In the studied cases, 36 patients (90%) had heterogeneously enhancing lesion, two patients (5%) had marginally enhancing lesions, one patient (2.5%) had homogenously enhancing lesion, and one patient (2.5%) had a nonenhancing lesion (χ2 = 33.270; P value = 0.001).
Twenty-seven cases were diagnosed as primary tumors (16 high grade and 11 low grade), three cases were diagnosed as metastasis, six cases were diagnosed as primary versus metastatic, and four cases were diagnosed as non-neoplastic lesions [Table 2].
MRS evaluation of the 40 patients revealed that 31 of the 40 patients (77.5%) had primary tumors, five patients (12.5%) had metastatic tumors, and four cases had non-neoplastic lesions (10%) [Table 3] and [Figure 4].
|Figure 4: Case 4: An MRI showing multiple, well-defi ned, cystic space occupying lesions seen in both cerebellar and paraventricular regions, both temporal and both parietal regions, displaying low signals on T1WIs (a) and high signals on T2WIs and fl air (b, c). An MRS showing intralesional voxels of long TE 144 (d) and short TE (e). It showed an increase in Cho, lipids, and lactate, with decrease in creatine and NAA and a Cho/NAA ratio of 5.6 and a Cho/Cr ratio of 4.9. Note: perilesional voxel show essentially normal metabolites. The above MRI and MRS are in favor of multiple metastatic lesions|
Click here to view
|Table 3 The space occupying lesion groups according to the MRS evaluation|
Click here to view
The calculated Cho/Cr ratios from the intralesional areas showed a significant increase from low-grade to high-grade tumors, with no significant difference between high-grade primary and metastases [Table 4] and [Figure 5].
|Figure 5: Case 5: An MRI showing bilateral, supratentorial, numerous, scattered, space occupying lesions that are small, rounded, and illdefined, and the largest one is at the left occipital lobe seen displaying low signals on T1WIs (a) and high signals on T2WIs and flair (b, c). They showed heterogeneous enhancement after intravenous contrast (d). An MRS showing intralesional voxels in long TE 144 (e) with a Cho/NAA ratio of 1.7 and a Cho/Cr ratio of 2 and increased lactate in short TE 35 (g). The above MRI and MRS are in favor of active multiple sclerosis|
Click here to view
The calculated Cho/Cr ratios from the perilesional areas showed an increase in Cho/Cr ratios (>1) in primary high-grade tumors (indicating perilesional infiltration), whereas there was no increase in Cho/Cr ratio (≤1) in cases of metastasis (indicating no perilesional infiltration) and non-neopastic lesions [Table 5] and [Figure 6].[Figure 7]
|Figure 6: Case 6: An MRI showing two space occupying lesions, with the largest in the left thalamic, posterior limb of the internal capsule and the smaller one seen in the right globus pallidus, displaying hypointense signals on T1WIs (while the lesion in right globus pallidus showing thin peripheral hyperintense ring in T1WIs) (a) and hyperintense signals on T2WIs and flair (b, c), with mild enhancement after intravenous contrast (d). MRI suspected astrocytoma versus neurofibromatosis for MRS. An MRS showing intralesional voxels of long TE 144 (e) and short TE 35 (g). It showed a Cho/NAA ratio of 1.1 and a Cho/Cr ratio of 1.4. Perilesional voxels of long TE 144 (f) showed a Cho/NAA ratio of 1. The above MRI and MRS are in favor of Neurofi bromatosis|
Click here to view
|Figure 7: Case 7: An MRI showing intra-axial, well-defined, space occupying lesion in the right temporoparietal regions, displaying low signals on T1WIs (a) and high signals on T2WIs and flair (b, c), with central areas of breakdown and heterogeneous enhancement after intravenous contrast injection (d) with surrounding grade II vasogenic edema. MRI is suggestive ofGBM. An MRS showing intralesional voxel of long TE 144 (e) with increased Cho, lipids, and lactate and decreased creatine and NAA, with a Cho/Cr ratio of 3 and a Cho/NAA ratio of 3.5. Perilesional voxels (f) showed increased Cho, lipids, and lactate and decreased NAA and Cr (Cho/Cr ratio = 4). The above MRS and MRI are in favor of high-grade primary brain tumor with perilesional infiltration. Histological examination revealed glioblastoma multiform. MRS, magnetic resonance spectroscopy; NAA, N-acetyl-l-aspartate|
Click here to view
Lipid and lactate levels
Low-grade primary tumors showed low lactate and lipid peaks, and, with increasing malignancy, tumors showed increasing levels of lactate and lipid peaks (indicating necrosis) with remarkable difference in lipid peaks between low and high grade tumors. There was no significant difference between primary and metastatic brain tumors as regards lactate peak.
Final histopathological results
The final histopathological results of 36 patients with neoplastic lesions revealed 31 cases of primary tumors (77.5%), which were as follows: 11 cases of low-grade glioma (27.5%), three cases of anaplastic astrocytoma (7.5%), 13 cases of glioblastoma multiformis (32.5%), one case of gliomatosis cerebri (2.5%), one case of oligodendroglioma (2.5%), and two cases of lymphoma (5%). There were five cases of metastasis (12.5%), and the other four cases were of non-neoplastic lesions (10%).
| Discussion|| |
Conventional MRI is still the basic imaging requirement for initial evaluation of brain tumors. It gives basic information about the anatomical features of these tumors, such as edema, mass effect, and pattern of enhancement  .
MRS limits the use of invasive diagnostic approaches such as brain biopsy, as histopathology is the gold standard for the diagnosis of brain tumors  .
Proton MRS may enable differentiation between lesions showing similar aspects on conventional MRI. Its accuracy is 95-100% in distinguishing neoplastic and non-neoplastic lesions, increasing diagnostic specificity when associated with conventional MRI. Cho is considered the most specific marker of intracranial neoplasm. Increase in Cho levels and Cho/Cr and Cho/NAA ratios is highly suggestive of neoplasm; this is in agreement with the findings of Brandαo and Domingues  .
In contrast to the structural information provided by MRI, MRS provides a qualitative analysis of a number of metabolites within the brain. These metabolites reflect aspects of neuronal integrity, cell membrane proliferation or degradation, energy metabolism, and necrotic transformation of the brain or tumor tissue  .
All cases in our study were subjected to MRI examination, for which the nature of the lesions was doubtful. Thus, MRS examination was recommended to be used as a complementary tool for reaching the accurate diagnosis. It was important to detect the viable component of the lesion in the postcontrast study, where this part of the lesion was the one to be examined - that is, the voxel was placed at this part. The size of the lesion was another important factor, as accordingly the use of either the single or the multivoxel technique was planned.
In current study, 40 patients were examined in the Diagnostic Radiology and Medical Imaging Department, Tanta University, after neurological examination. Patients were studied with MRS with long and short TE to evaluate metabolic pattern of lesions.
In our study, the limitations of MR spectroscopy were the patient's involuntary movements and irritability, decreasing the sensitivity and specificity of the examination, as the position of the voxel was altered. This caused false results, and so repeating the examination was essential, leading to more time consumption. This difficulty was overcome, by reassuring the patient or even sedating.
In 2004, Brandαo and Domingues  stated that, to have high-quality spectra, blood, blood products, air, cerebrospinal fluid, fat, necrotic areas, metal calcification, and bone must be avoided. In such areas, differing magnetic susceptibility results in a nonhomogenous field that hinders the production of diagnostic quality spectra.
The important ratios that were calculated were Cho/NAA (within the lesion) and Cho/Cr (both within the lesion and normal side) ratios, where Cho/NAA and Cho/Cr were seen elevated in the neoplastic lesions. NAA/Cr ratio was not a sensitive or specific value.
Correlation between MR appearance of the lesions and the MRS findings was important in giving the most accurate suggested diagnosis.
This study was conducted on 40 patients, 24 male (60%) and 16 female (40%) patients. This is in agreement with the findings of Al-Okaili et al.  , who found that the incidence of brain tumors among male is more than that in female patients.
This sample represented a wide range of age group ranging from 8 to 66 years, with the majority of our cases ranging from 40 to 50 years in 12 cases (30 %). This is in agreement with the findings of Lehnhardt et al.  , who reported that the peak age incidence lies between 40 and 50 years.
In this study, 28 patients (70%) complained of headache, 20 patients (50%) had cerebral dysfunction (disturbed conscious level and generalized convulsions), 12 patients (30%) were suffering from hemiplegia or hemiparesis, four patients (10%) had visual complaints (diminution of vision or diplopia), and two patients (5%) had behavioral changes. These results are in agreement with those of Patel  , who reported that headache is the most common presenting symptom, followed by seizures and hemiparesis.
As regards the distribution of the lesions, it was found that 18 patients (45%) had lesions in the parietotemporal regions, 12 patients (30%) had lesions in the occipital regions, eight patients (20%) in the frontal regions, and two patients (5%) had lesions that were invading and crossing mid line through the corpus callosum. This is in agreement with the findings of Patel  .
In the studied cases, as regards the pattern of enhancement after gadolinium injection, there were 35 patients (87.5%) who had heterogeneously enhancing lesions, two patients (5%) with marginally enhancing lesions, one patient (2.5%) with homogenously enhancing lesion, and two patients (5%) with nonenhancing lesion. This is in agreement with the findings of Young and Knopp  , who also stated that contrast can help improve specificity and reflect break in blood-brain barrier.
As regards the conventional MRI evaluation for the studied cases, it diagnosed 27 cases as primary tumors (16 high grade and 11 low grade), three cases were diagnosed as metastasis, six cases were diagnosed as primary versus metastasis, and four cases were diagnosed as non-neoplastic lesions.
Eleven cases of low-grade primary tumors were surrounded by edema and none showed hemorrhage or necrosis.
In the 16 cases of high-grade primary tumors, eight of them showed hemorrhage, 13 showed necrosis, and 16 of them were surrounded by edema.
In the three cases of metastasis, two of them showed necrosis, three of them were surrounded by edema, and no cases showed hemorrhage. This is in agreement with the findings of Young and Knopp  , who stated that the presence of hemorrhage suggests certain primary and secondary tumors, including metastatic renal cell carcinoma, melanoma, and breast carcinoma.
Four cases of non-neoplastic lesions did not show hemorrhage, necrosis, or edema.
On long TE, the important metabolites that had to be commented upon their peaks, Cho,NAA, Cr. However, MI and lipid/lactate peaks were better to be commented upon on short TE.
The MRS results showed that all neoplastic lesions showed variable degrees of increased Cho peak and Cho/Cr ratios, with a significant increase from low-grade and high-grade tumor without any significant difference between primary and metastatic brain tumors. This is in agreement with the results of Martinez-Bisal and Celda  and Shokry  , who found that an increase in Cho/Cr ratio in lesion could only differentiate low-grade primary tumors from high grade.
In our study, the MRS examination was of great benefit in differentiating neoplastic from non-neoplastic lesions, which helps in the subsequent management and saving some patients the need for biopsy taking.
Cho was the main metabolite to be assessed in neoplasms. Increased Cho levels are associated with higher cell membrane turnover and higher cell density arising from the proliferation of tumor cells  .
Typically, the higher levels of Cho occur in grade II and III gliomas; however, in glioblastoma multiforme (GBM), the levels could be much lower due to necrosis. In fact, Cho may be reduced in GBM when the spectrum is acquired in the area of tumor necrosis. It is important to include as much solid lesion as possible in the voxel, to show increased Cho levels while avoiding regions of necrosis. An increase in Cho/Cr and Cho/NAA ratios should suggest neoplasm  .
A decreased Cr peak was also detected in all neoplastic cases, but without significant difference between primary and metastatic brain tumors. This is in agreement with the results of Delorme and Weber  , who stated that, in a tumor, the increase in cell turnover will lead to an increase in Cho concentration, along with a depression of the NAA peak due to the loss in healthy glioneural structures. The creatine resonance may be depressed to varying degrees (depending on the energy status of the tumor) and often serves as an internal reference for calculating ratios. When there are hypoxic components, a lactate resonance may be present. A lipid peak indicates the presence of necrosis and suggests a high grade of malignancy.
In short TE (35 ms), the lactate levels showed a significant difference between low-grade and high-grade tumors as well, without any significant difference between primary and metastatic brain tumors or between low-grade tumors and non-neoplastic lesions. This is in agreement with the results of Van der Graaf  , who stated that the presence of the lactate peaks is usually consistent with aggressive tumors, reflecting increased anaerobic metabolism and cellular necrosis, and this pathology does not differ a lot from primary to metastatic brain tumors.
Short TE 35 showed the higher lipid peaks in metastatic brain tumors and showed significant difference in the levels from primary lesions. This is in agreement with the findings of Shokry  , Opstad et al.  , and Van der Graaf  , who gave a possible explanation for elevated lipids in metastatic lesions as the cancer cells of different origin contain mobile, spectroscopically detectable lipids in their cell membrane, whereas cell cultures of tumors that do not lead to metastases do not exhibit these mobile lipids.
In this study, intralesional voxels in primary tumors showed different levels of Cho/Cr ratios, which showed an increase in Cho/Cr ratios with the increase in the grade of the tumor. Therefore, Cho/Cr ratios have shown consistency in predicting the tumor grade; the comparison of Cho/Cr ratios in high-grade tumors with that of low-grade tumors proved a significant difference between the two groups.
Chen et al.  and Chiang et al.  added that the increase in Cho/Cr ratios in high-grade tumors than in low-grade tumors was significantly correlated with the expression of proliferating cells.
There was increase in perilesional Cho/Cr ratio in high-grade primary tumors, with significant level differences from metastatic brain tumors and non-neoplastic lesions, which did not show an increase in perilesional Cho/Cr ratio. This is in agreement with the findings of Faria et al.  , who added that primary high-grade tumors had been reported to have peritumoral infiltrating neoplastic cells. Thus, perilesional edema showed spectroscopic malignant changes in the form of higher Cho/Cr ratio in primary tumors compared with metastatic and non-neoplastic lesions. Moreover, Cr peaks showed evident decreased levels in perilesional area of high-grade primary tumors with associated higher Cho levels than that in perilesional areas of metastatic brain tumors.
As regards the specific MRS appearance of each lesion, it was mentioned that low-grade gliomas may also show marked elevation of MI (Myo-inositol), whereas high-grade gliomas show normal or no MI. Oligodendrogliomas have spectroscopic characteristics similar to high-grade astrocytoma, with high levels of Cho, but may also have elevated MI.  .
In GBM cases, the MI was absent, whereas pathologically proven cases of grade II and III glial tumors showed moderate-to-high MI peaks. However, in oligodendroglioma, in addition to the elevated Cho level, the MI and lipid/lactate peaks were elevated as well.
In 2000, Maheshwari et al.  mentioned that cortical dysplasia can be confused with low-grade glioma in routine MRI scans. MRS can differentiate between the two by showing raised Cho peaks in tumors. Cortical dysplasia may show some reduction in NAA levels, but it is not as marked as in tumors.
In 1993, Castillo et al.  stated that lymphomas show reduction of NAA and high Cho, similar to astrocytomas. Lymphoma may also show infiltration of adjacent brain, and may be indistinguishable from astrocytoma. In a reported case, the level of NAA was nearly normal, and there was only minimal elevation of Cho.
Sometimes, conventional MRI alone cannot solve problem. MRS, therefore, could help in their differentiation. Increased Cho peak and increased Cho/Cr ratio were the two main abnormalities that manifested themselves in this study, which included various types of intracranial space occupying neoplastic lesions. All these cases showed increased Cho peak and Cho/Cr ratio in intralesional voxels, with a significant increase from low-grade to high-grade tumors.
Metastases showed large lipid peak even in the absence of necrosis, with the absence of elevated Cho peak and Cho/Cr ratio in the surrounding peritumoral edema.
Plaques of multiple sclerosis are characterized by lactate increase, depending on the degree of inflammatory reaction, and by increased Cho/Cr ratios due to acute myelin breakdown, but usually not as intense as in neoplastic lesions. Chronic plaques present reduced NAA/Cr ratios in the center of the lesion.
| Conclusion|| |
Conventional MRI has been the long-established method for evaluating brain tumors before and after treatment.
MRS gives information about the biochemical changes in tissues, which appear earlier than the structural changes, and so, in recent times, MRS has been used as a noninvasive method for the diagnosis and grading of brain tumors.
Intralesional Cho/Cr ratio could not differentiate primary tumors from metastasis but it was able to grade tumor and differentiate between high-grade and low-grade tumors, with no significant difference between low-grade tumors and non-neoplastic lesions.
The presence of lipid and lactate also denotes high-grade tumors. The lactate levels showed a significant difference between low-grade and high-grade tumors, without any significant difference between primary and metastatic brain tumors.
Perilesional Cho/Cr ratio could differentiate primary tumors from metastasis. Metastasis exhibits prominent lipid level even in the absence of much necrosis.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Tilgner J, Herr M, Ostertag C, Volk B. Validation of intraoperative diagnoses using smear preparations from stereotactic brain biopsies: intraoperative versus final diagnosis - influence of clinical factors. Neurosurgery 2005; 56: 257-265.
Aasly J, Silfvenius H, Aas TC, Sonnewald U, Olivecrona M, Juul R, White LR. Proton magnetic resonance spectroscopy of brain biopsies from patients with intractable epilepsy. Epilepsy Res 1999; 35:211-217.
Abeloff MD, Armitage JO, Lichter AS. Clinical oncology. 3rd ed. Edinburgh, UK: Churchill Livingston; 2004. 20. 72-77.
Adamson AJ, Rand SD, Prost RW, Kim TA, Schultz C, Haughton VM. Focal brain lesions: effect of single-voxel proton MR spectroscopic findings on treatment decisions. Radiology 1998; 209:73-78.
Ando K, Ishikura R, Nagami Y, Morikawa T, Takada Y, Ikeda J, et al
. Usefulness of Cho/Cr ratio in proton MR spectroscopy for differentiating residual/recurrent glioma from non-neoplastic lesions. Nihon Igaku Hoshasen Gakkai Zasshi 2004; 64:121-126.
Angelie E, Bonmartin A, Boudraa A, Gonnaud PM, Mallet JJ, Sappey-Marinier D. Regional differences and metabolic changes in normal aging of the human brain: proton MR spectroscopic imaging study. Am J Neuroradiol 2001; 22:119-127.
Shokry A. MRS of brain tumors: diagrammatic representations and diagnostic approach. Egypt J Radiol Nucl Med 2012; 43:603-612.
Burger PC, Scheithauer BW. Pilocytic astrocytoma In: System, Kleihues P, Cavenee WK editors. Pathology and genetics of tumours of the nervous
. Lyon, France: International Agency for Research on Cancer; 2000. 45-51.
Brandáo LA, Domingues RC. MR spectroscopy of the brain
. Philadelphia: Lippincott Williams & Wilkins; 2004. 5. 33-85.
Sibtain NA, Howe FA, Saunders DE. The clinical value of proton magnetic resonance spectroscopy in adult brain tumours. Clin Radiol 2007; 62: 109-119.
Al-Okaili RN, Krejza J, Wang S, Woo JH, Melhem ER. Advanced MR imaging techniques in the diagnosis of intra axial brain tumors in adults. Radiographics 2006; 26:S173-S189.
Lehnhardt FG, Bock C, Röhn G, Ernestus RI, Hoehn M. Metabolic differences between primary and recurrent human brain tumors: a 1H NMR spectroscopic investigation. NMR Biomed 2005; 18:371-382.
Patel AM. Brain tumor. Magnetic resonance spectroscopy study of metabolic profiles in brain tumors USA: ProQuest LLC; 2010. 2. 10-18.
Young RJ, Knopp EA. Brain MRI: tumor evaluation. J Magn Reson Imaging 2006; 24:709-724.
Martinez-bisbal MC, Celda B. Proton MR spectroscopy imaging in the study of human cancer. J Nucl Med Mol Imaging 2009; 53:618-630.
Delorme S, Weber MA. Applications of MRS in the evaluation of focal malignant brain lesions. Cancer Imaging 2006; 6:95-99.
van der Graaf M. In vivo magnetic resonance spectroscopy: basic methodology and clinical applications. Eur Biophys J 2010; 39: 527-540.
Opstad KS, Murphy MM, Wilkins PR, Bell BA, Griffiths JR, Howe FA. Differentiation of metastases from high-grade gliomas using short echo time 1H spectroscopy. J Magn Reson Imaging 2004; 20:187-192.
Chen J, Huang SL, Li T, Chen XL. In vivo research in astrocytoma cell proliferation with 1H-magnetic resonance spectroscopy: correlation with histopathology and immunohistochemistry. Neuroradiology 2006; 48: 312-318.
Chiang IC, Kuo YT, Lu CY. Distinction between high-grade gliomas and solitary metastases using peritumoral 3-T magnetic resonance spectroscopy, diffusion, and perfusion imagines. Neuroradiology 2004; 46:619-627.
Faria AV, Macedo FC, Marsaioli AJ. Classification of brain tumor extracts by high resolution 1
H MRS using partial least squares discriminate analysis. Braz J Med Biol 2011; 44:1678-4510.
Castillo M, Smith JK, Kwock L. Correlation of myo-inositol levels and grading of cerebral astrocytomas. Am J Neuroradiol 2000; 21: 1645-1649.
Maheshwari SR, Fatterpekar GM, Castillo M, Mukherji SK. Proton MR spectroscopy of the brain. Semin Ultrasound CT MR 2000; 21:434-451.
Castillo M, Kwock L, Scatliff J, Gudeman S, Greenwood R. Proton MR spectroscopic characteristics of a presumed giant subcortical heterotopia. Am J Neuroradiol 1993; 14:426-429.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]