|
 
 |
| ORIGINAL ARTICLE |
|
| Year : 2022 | Volume
: 29
| Issue : 1 | Page : 33-41 |
|
Multiparametric differentiation of intracranial central nervous system lymphoma and high-grade glioma using diffusion-, perfusion-, susceptibility-weighted magnetic resonance imaging, and spectroscopy
Santosh Rai1, Fathima Raeesa1, Mayur Kamath2, Sharada Rai3, Muralidhar K Pai2, Sonali D Prabhu1
1 Department of Radiodiagnosis, Kasturba Medical College, Mangalore, Manipal Academy of Higher Education, Manipal, Karnataka, India
2 Department of Neurosurgery, Kasturba Medical College, Mangalore, Manipal Academy of Higher Education, Manipal, Karnataka, India
3 Department of Pathology, Kasturba Medical College, Mangalore, Manipal Academy of Higher Education, Manipal, Karnataka, India
| Date of Submission | 03-May-2021 |
| Date of Acceptance | 01-Jul-2022 |
| Date of Web Publication | 15-Nov-2022 |
Correspondence Address:
Dr. Santosh Rai
Department of Radiodiagnosis, Kasturba Medical College, Manipal Academy of Higher Education, Mangalore - 575 001, Karnataka
India
 Source of Support: None, Conflict of Interest: None  | Check |
DOI: 10.4103/wajr.wajr_16_21
Aims and Objectives: To observe the characteristics of primary central nervous system lymphoma (PCNSL) and high-grade glioma (HGG) in diffusion-weighted imaging (DWI), perfusion-weighted imaging (PWI), susceptibility-weighted imaging (SWI) and spectroscopy, and differentiate them based on these parameters.
Materials and Methods: A total of 45 patients diagnosed with the central nervous system (CNS) neoplasm on magnetic resonance imaging (MRI) using 1.5 Tesla MRI Siemens Magnetom Avanto (Siemens, Germany) and with subsequent histopathological evidence as glioblastoma or CNS lymphoma were included. The study was completed over a period of 2 years.
Results: It was found that DWI is effective in the differentiation of HGGs and PCNSLs. A total of 20 (57.1%) HGGs showed diffusion restriction, whereas 9 (90%) of the PCNSLs showed diffusion restriction. The mean apparent diffusion coefficient (ADC) (×10–6 mm2/s) in PCNSLs was 646 whereas, in HGGs, it was found to be 824.3. Thirty-one (88.6%) of the HGGs showed increased perfusion. The mean value of rCBVmean in HGG was found to be 4.06 and the mean value of rCBVmax was 3.63. None of the PCNSLs showed increased perfusion. The mean value of rCBVmean in PCNSLs was 0.097 and rCBVmax was 0.133. 30 (85.7%) of HGGs showed significant areas of blooming on SWI (in the form of intratumoral susceptibility signals [ITSS]). None of the PCNSLs showed blooming. All HGGs and PCNSLs showed increased choline and decreased N acetyl aspartate (NAA) on spectroscopy. Mean Choline/Creatine (Cho/Cr) in HGGs was found to be 3.06, whereas in PCNSLs, it was 1.84.
Conclusion: It is important to make a distinction between HGG and PCNSL as the treatment modalities are different for both. Multiparametric evaluation of ADC, ITSS, and rCBVmean allows the differentiation of PCNSLs and solid glioblastoma which supports the integration of advanced MR imaging techniques including DSC-PWI, DWI, and SWI for the routine diagnostic workup of these tumors.
Keywords: Diffusion-weighted imaging, high-grade glioma, multiparametric magnetic resonance imaging, perfusion-weighted imaging, primary central nervous system lymphoma, spectroscopy, susceptibility-weighted imaging
How to cite this article:
Rai S, Raeesa F, Kamath M, Rai S, Pai MK, Prabhu SD. Multiparametric differentiation of intracranial central nervous system lymphoma and high-grade glioma using diffusion-, perfusion-, susceptibility-weighted magnetic resonance imaging, and spectroscopy. West Afr J Radiol 2022;29:33-41 |
How to cite this URL:
Rai S, Raeesa F, Kamath M, Rai S, Pai MK, Prabhu SD. Multiparametric differentiation of intracranial central nervous system lymphoma and high-grade glioma using diffusion-, perfusion-, susceptibility-weighted magnetic resonance imaging, and spectroscopy. West Afr J Radiol [serial online] 2022 [cited 2023 Mar 31];29:33-41. Available from: https://www.wajradiology.org/text.asp?2022/29/1/33/361182 |
| Introduction | |  |
There is an increasing prevalence of central nervous system (CNS) tumors in India, with the incidence ranging from 5 to 10/100,000 population. The CNS tumors account for 2% of all malignancies.[1]
It is important to noninvasively differentiate between primary central nervous lymphoma (PCNSL) and high-grade glioma (HGG) as the treatment modalities are different for both. Gross total resection, followed by chemoradiation therapy is the treatment of choice in HGGs. However, patients with PCNSL require stereotactic biopsy and then chemotherapy.[2]
Conventional magnetic resonance imaging (MRI) is not sufficient in itself for the differentiation of PCNSL and HGGs, and tissue diagnosis is essential before the commencement of treatment. Therefore, we need other modalities for their differentiation such as diffusion-weighted imaging (DWI), perfusion-weighted imaging (PWI), susceptibility-weighted imaging (SWI), and magnetic resonance (MR) spectroscopy.
The basis of DWI is the random Brownian motion More Details of water molecules within tissues. Apparent diffusion coefficient (ADC) can be calculated from DWI sequence, which inversely correlates with tumor cell density.[2]
Information about cerebral physiology at the capillary level is provided by PWI. Among a few PWI techniques, dynamic susceptibility contrast (DSC) MR imaging is the most often used. Cerebral blood volume (CBV) maps and noninvasive measurements of relative CBV (rCBV) are provided by DSC MRI. rCBV parameter correlates with tumor vascularity and thus is increased in tumors with a high rate of pathologic neoangiogenesis.[2]
SWI is particularly sensitive to compounds that distort the local magnetic field. It is useful in detecting blood products and calcium, which are usually readily identifiable in conventional MRI. Intratumoral susceptibility signals (ITSS) are noted within the lesions.
MR spectroscopy assesses biochemical changes and relies on the chemical shift phenomena of proton (hydrogen) nuclei.[2] It detects the presence and concentration of various metabolites in tissues, thereby improving the ability to predict histological grade.
Multiple studies have been done in the past to differentiate between intracranial CNS lymphomas and HGGs, but there are only a handful of studies using multiparametric MRI combining DWI, PWI, SWI, and spectroscopy. The aim of our study is to determine the degree which multiparametric MRI could differentiate between PCNSL and GB. The objective was to acquire the ADC values of HGG and intracranial CNS lymphoma using DWI; to acquire rCBV values of HGG and intracranial CNS lymphoma using DSC PWI; to observe ITSS of HGG and intracranial CNS lymphoma in SWI and to note the peaks of various metabolites in HGG and intracranial CNS lymphoma using MR spectroscopy. Using the above parameters to differentiate between HGG and intracranial CNS lymphoma.
| Materials and Methods | | |
The study was conducted after obtaining permission from the institutional ethics committee. This was an observational study and data were collected over 2 years from September 2018 to August 2020 from KMC Hospital, Ambedkar Circle, Mangalore. Patients who were subjected to MRI and were found to have a CNS neoplasm with subsequent histopathological evidence such as glioblastoma or CNS lymphoma were included in the study. We obtained samples from 45 patients. ADC from DWI, rCBV from PWI, ITSS from SWI, and metabolite peaks from MR spectroscopy was recorded. Siemens Magnetom Avanto 1.5 Tesla (Siemens, Germany) MRI machine was used with the following parameters as shown in [Table 1]. | Table 1: Siemens Magnetom Avanto 1.5 Tesla (Siemens, Germany) magnetic resonance imaging machine was used with the following parameters
Click here to view |
DWI was performed with a single-shot spin-echo echo-planar sequence with the above parameters. Diffusion sensitizing gradients were applied sequentially in the x, y, and z directions with b = 0 and 1000 s/mm2.
DSC-PWI was performed with a T2*-weighted gradient-echo echo-planar imaging sequence with the above parameters. Dynamic contrast (Gadolinium) 10 ml followed by flush was given at an injection rate of 2.5 ml/s.
Spectroscopy was performed using multivoxel spectroscopy versus chemical shift imaging. Frequency domain curve as fitted by the manufacturer to define N-acetyl aspartate (NAA), Choline-containing components (Cho), Creatine and Phosphocreatine (Cr), and lipid and lactate peaks.
All patients were subjected to biopsy and the diagnosis was confirmed by the Department of Pathology at KMC, Mangalore. Data analysis was done using the Chi-square test. SPSS software (version 22.0, IBM SPSS software, Mangalore, Karnataka, India) was used to do the analysis.
| Results | | |
We studied the MRI appearance (tumor sizes, edema, contrast enhancements, multiplicity, and laterality) of histologically confirmed cases of HGGs (Grade III and Grade IV) and primary CNS lymphomas and tried to distinguish them based on DWI, PWI, SWI, and spectroscopy.
A total of 45 patients were eligible, of which there were 35 patients with HGGs and 10 patients with PCNSL. Of the 35 HGG patients, 9 (25.7%) were female participants and 26 (74.3%) were male participants. Of the PCNSLs, only 1 (10%) participant was female.
None of the patients with CNS lymphoma were immunocompromised.
The mean age of subjects with HGGs was 53.11 ± 14.305 years and the median age was 55 years, whereas in PCNSLs, the mean age was 53 ± 8.832 years and median age was 53 years.
The majority of subjects with HGGs were in the age group of 51–60 years (28.6%).
A total of 28 (80%) of HGGs were located in the cerebral hemispheres (frontal, temporal, occipital, and parietal), whereas 3 (8.6%) were found in the corpus callosum and periventricular locations. One (2.9%) HGG was in an unusual location, the cerebellum. In the case of PCNSLs, 4 (40%) were located in the lobar region, whereas 1 (10%) was in the brainstem, 1 (10%) in the choroid plexus, 1 (10%) in the corpus callosum, 1 (10%) in the cerebral lobe as well as in the corpus callosum, and 2 (20%) were periventricular in location. The choroid plexus and middle cerebellar peduncle were the uncommon locations for PCNSL in our study. Representative cases of HGG (Grade III and Grade IV) are depicted in [Figure 1],[Figure 2],[Figure 3],[Figure 4],[Figure 5],[Figure 6]. Representative cases of Lymphoma are depicted in [Figure 7] and [Figure 8]. The corresponding intraoperative images are depicted in [Figure 5] and [Figure 8]. | Figure 1: (a and b) Axial T1 and T2 images showing solid cystic lesion in the right frontal lobe (white arrows). (c) Axial post contrast T1 image showing peripheral as well as a solid enhancement (white arrow). (d) SWI showing multiple foci of blooming (white arrow). (e) PWI showing increased perfusion in the solid portion (white arrow). (f) DWI showing diffusion restriction in the solid component (white arrow). (g) Spectroscopy showing elevated choline and reduced NAA in the lesion. (h) Hematoxylin and eosin stain showing infiltrative cellular high-grade glial neoplasm composed of fibrillary and pleomorphic astrocytes along with multinucleate tumor giant cells on a gliofibrillary stroma – Glioblastoma, NOS, WHO Grade IV, SWI - Susceptibility-weighted imaging, PWI - Perfusion-weighted imaging, DWI - Diffusion-weighted imaging, NAA - N acetyl aspartate, NOS - Nitric oxide synthase
Click here to view |
 | Figure 2: (a and b) Axial T1 and T2 images showing an ill-defined solid lesion in the left parieto-occipital lobe (white arrows). (c) Axial postcontrast T1 image showing patchy peripheral enhancement (white arrow). (d) SWI showing multiple blooming foci.(white arrow). (e) PWI not showing increased perfusion (within the circle) (white arrow). (f) DWI showing diffusion restriction (white arrow). (g) Spectroscopy showing elevated choline and reduced NAA in the lesion. (h) Hematoxylin and eosin stain showing moderately cellular glial neoplasm composed of fibrillary, pleomorphic, gemistocytic, and multinucleate astrocytes on a gliofibrillary stroma – Glioblastoma (NOS) – WHO Grade IV, SWI - Susceptibility-weighted imaging, PWI - Perfusion-weighted imaging, DWI - Diffusion-weighted imaging, NAA - N acetyl aspartate, NOS - Nitric oxide synthase
Click here to view |
 | Figure 3: Intraoperative image of Glioblastoma as in [Figure 2], (a) Pre-excision shows a grayish mass with areas of hemorrhage and necrosis in the left parieto-occipital region arising from subcortical white matter, which is soft and friable with poorly defined mass and extensive infiltration (black arrow). (b) Image shows postexcision cavity (black arrow)
Click here to view |
 | Figure 4: (a and b) Axial T1 and T2 images showing a lobulated predominantly cystic lesion in the right parietal lobe (white arrows). (c) Axial post contrast T1 image showing peripheral enhancement (white arrow). (d) SWI showing few blooming foci (white arrow). (e) PWI showing increased perfusion (white arrow), (f) DWI showing peripheral diffusion restriction (white arrow) (g) Spectroscopy showing elevated choline and reduced NAA in the lesion. (h) Hematoxylin and eosin stain showing fragments of gliotic brain tissue with areas of hemorrhage, increased cellularity, pleomorphism, and occasional mitosis – Glioblastoma– WHO Grade IV, SWI - Susceptibility-weighted imaging, PWI - Perfusion-weighted imaging, DWI - Diffusion-weighted imaging, NAA - N acetyl aspartate, NOS - Nitric oxide synthase
Click here to view |
 | Figure 5: Intraoperative image of glioblastoma as in [Figure 4], (a) Preexcision shows an intraparenchymal tumor in the right parietal lobe consisting of soft and friable tissue, poorly defined, with areas of hemorrhage and central necrosis and surrounding extensive infiltration, (black arrow) (b) Image shows postexcision cavity (black arrow)
Click here to view |
 | Figure 6: (a and b) Axial T1 and T2 images showing a well-defined solid lesion in the right medial temporal lobe (white arrows). (c) Axial post contrast T1 image showing peripheral enhancement (white arrow). (d) SWI showing no blooming foci (white arrow). (e) PWI showing increased perfusion (white arrow). (f) DWI showing no diffusion restriction (white arrow). (g) Spectroscopy showing elevated choline and reduced NAA in the lesion. (h) Hematoxylin and eosin stain showing neoplastic astrocytes exhibiting mild to moderate degree of nuclear atypia and scattered mitotic figures– Glioblastoma (NOS)– WHO Grade IV, SWI – Susceptibility-weighted imaging, PWI – Perfusion-weighted imaging, DWI – Diffusion-weighted imaging, NAA – N acetyl aspartate, NOS – Nitric oxide synthase
Click here to view |
 | Figure 7: (a and b) Axial T1 and T2 images showing a solid lesion in the right thalamus (white arrows), (c) Axial postcontrast T1 image showing homogeneous enhancement (white arrow), (d) SWI showing no blooming foci (white arrow) (e) PWI showing no increased perfusion (within the circle) (white arrow) (f) DWI showing diffusion restriction in the lesion (white arrow) (g) Spectroscopy showing elevated choline and reduced NAA in the lesion. (h) Hematoxylin and eosin stain showing fragments of neuroparenchyma along with fragments of highly cellular neoplasm with hyperchromatic nuclei, scant eosinophilic cytoplasm and prominent angiocentricity– Non-Hodgkin's lymphoma, SWI – Susceptibility-weighted imaging, PWI – Perfusion-weighted imaging, DWI – Diffusion-weighted imaging, NAA – N acetyl aspartate
Click here to view |
 | Figure 8: Case of primary CNS lymphoma, as in [Figure 7] (a) Preexcision shows a well-circumscribed lesion in the right periventricular region, which is firm, gray to tan yellowish, with areas of hemorrhage and no necrosis, (b) Image shows postexcision cavity, CNS – Central nervous system
Click here to view |
Diffusion-weighted imaging results
The mean ADC (×10–6 mm2/s) in PCNSLs was 646, whereas, in HGGs, it was found to be 824.3. The lesions appeared more hyperintense on the B series of lymphoma with lower ADC values [Figure 7] [Table 2]a.
Perfusion-weighted imaging results
A total of 31 (88.6%) HGGs showed increased perfusion [Figure 1], [Figure 2], [Figure 4], [Figure 6]. The mean value of rCBVmean in HGGs was found to be 4.06 and the mean value of rCBVmax was 3.63. None of the PCNSLs showed increased perfusion. The mean value of rCBVmean in PCNSLs was 0.097 and rCBVmax was 0.133 [Table 2]b.
Susceptibility-weighted imaging results
A total of 30 (85.7%) HGGs showed blooming on SWI in the form of ITSS.
(Grade 0, No ITSS, Grade 1, 1–5 dot-like or linear ITSS, Grade II, 6–10 dot-like or fine linear ITSS, and Grade III, more than 11 ITSSs) The grading of ITSS was done as I, II, and III in 13% (4 of 31), 23% (7 of 31), and 66% (20 of 31) of those with HGGs. The mean ITSS for HGGs was 2.2. None of the PCNSLs showed blooming [Figure 7] [Table 2]c.
Magnetic resonance spectroscopy results
All cases of HGGs and PCNSLs showed elevated choline and reduced NAA on spectroscopy with the voxel placed within the solid enhancing component of the lesion. Mean Cho/Cr in HGGs was 3.06, whereas in PCNSLs it was 1.84. 48.5% (17 of 35) of the HGGs showed a lipid/lactate peak. A total of 30% (3 of 10) of the PCNSLs showed a lipid/lactate peak [Table 2]d.
| Discussion | | |
We evaluated multiparametric MRI using DWI, PWI, SWI, and MR spectroscopy to preoperatively differentiate between primary CNS lymphoma and HGG in a small cohort in the Indian population study, which may sometimes be difficult to achieve by conventional MRI sequences.
The mean age of subjects in our study was found to be lower compared to the study by Kickingereder et al. male predominance was observed in cases of HGG, which was consistent with the study by Kickingereder et al. However, our study had a female preponderance in case of PCNSL.
The principal study finding is that primary CNS lymphomas have lower ADC values, lower rCBVmean and rCBVmax, and lower ITSS grades compared to HGGs.
Valuable information about the cellularity of tissue is demonstrated by DWI. Tissue cellularity correlates inversely with ADC. In malignancies, the intracellular proportion is increased, so the diffusion becomes relatively more restricted. A significantly higher diffusion restriction with lower ADC was seen in intracranial CNS lymphomas compared to HGGs in our study, which were consistent with studies done by Kickingereder et al.,[2] Wang et al.,[3] Doskaliyev et al.,[4] and Yamashita et al.[5] Lower mean ADC values imply higher cellularity within the tumor, which is concordant with multiple previous studies. In the study by Nesta-Matuszewska et al.,[6] which tried to differentiate between glioblastoma multiforme, metastases, and primary CNS lymphomas, low ADC values in cases of PCNSLs were found. In addition, the diffusion and perfusion parameters in the peritumoral T2 hyperintense zone in these lesions were measured. Kayed et al.[7] used advanced neuroimaging for the diagnosis of PCNSL and found low ADC values in all PCNSL patients, ranging from 0.61 to 0.67 × 10–3 mm2/s with mean ADC being 0.62 ± 0.025 × 10–3 mm2/s. This finding has been consistent with our study. Fawzy et al.[8] concluded that DWI had higher sensitivity, specificity, and accuracy than contrast-enhanced MRI and MRS in the grading of gliomas. Likewise, Boonzaier et al.[9] measured diffusion and perfusion in the enhancing and nonenhancing components in patients with glioblastoma. Significantly higher ADC values in the nonenhancing components compared to the solid enhancing areas were discovered, which is again consistent with our conclusion of higher ADC values in HGGs. In the study by Makino et al.,[10] the minimum ADCs in PCNSLs were found to be significantly lower than those of GBMs.
Perfusion-weighted imaging is an important tool in the assessment of vascularity of lesions. Vascular proliferation is a pathological hallmark of GBM, whereas PCNSLs lack neoangiogenesis. It is quantified by rCBV, which is the ratio of blood volume within the lesion to that in the contralateral normal brain parenchyma. Authors of several studies such as Kickingereder et al.,[2] Wang et al.,[3] Toh et al.,[11] Schob et al.,[12] and Neska-Matuszewska et al.[6] have shown high rCBV in HGGs compared to PCNSLs, which is consistent with our study. Similarly, the ratios of rCBVmean and rCBVmax were appreciably lower in patients with PCNSL than in those with glioblastoma, according to Kickingereder et al.[2] Pretreatment rCBV was reduced in PCNSLs compared to GBMs and metastases, according to the study done by Mangla et al.[13] In another study by Makino et al., maximum rCBVs aided in distinguishing PCNSLs and GBMs with 97% sensitivity and 90.7% specificity.[10] The higher perfusion values of rCBVmean and rCBVmax in case of HGGs are justified by the theory of extensive neovascularization in HGGs as explained by Hardee et al.[14] The PCNSLs have lower perfusion because of angiocentric pattern of growth and absence of neovascularization.
The SWI helps in the grading of cerebral tumors by providing valuable information about both hemorrhagic and calcified foci inside the tumor and allows the assessment of internal angioarchitecture. HGGs show blooming in the form of different ITSS grades. Higher ITSS grades were found in HGGs compared to PCNSLs because of neovascularization and intralesional bleeding. Ding et al.[15] attempted to differentiate brain malignancies using SWI. They found that intralesional hemorrhage and intralesional vessels were more appreciable in non-PCNSLs than PCNSLs. The quantification of intralesional vessels was effective in the differentiation of PCNSLs from non-PCNSLs with a sensitivity, specificity, PPV, and NPV of 100.0%, 82.7%, 62.2%, and 100.0%, respectively. The study by Radbruch et al.[16] showed that the presence of ITSS reliably differentiates between glioblastoma and B-cell PCNSL and provides a firm basis for the diagnosis without needing much postprocessing work.
On MR spectroscopy, both PCNSL and HGGs showed elevated choline and decreased NAA in our study. Mean Cho/Cr in HGGs was found to be higher than in PCNSLs. Authors of several studies[17],[18] have shown an association between Cho levels and tumor grade in gliomas. Higher choline levels are associated with more aggressive and higher-grade tumors. However, certain other studies by Howe et al.,[19] have shown that Grade IV tumors have lesser choline compared to Grade III tumors. This could be due to more necrosis being present in Grade IV tumors which limits the amount of tissue where the voxel for MRS could be placed. Necrotic tissue is also associated with low metabolism and cellular turnover. This could be another factor that could result in a lower amount of choline in Grade IV tumors than grade III tumors. In the study by Kayed et al.,[7] the mean Cho/Cr in PCNSL was found to be 19.1, which is higher than the values found in our study.
We also found lipid/lactate peaks within the lesions under evaluation. According to Li et al.[20] lipid/lactate peak is seen in high-grade tumors. This finding has been consistent in our study, where we found 48.5% of HGGs and 30% of the PCNSLs showing a lipid/lactate peak. Lipid/lactate peak was also seen in PCNSL in a study by Kayed et al.[7] The study by Fawzy et al.[8] revealed high Cho/Cr and lipid/lactate peaks in HGGs. They showed a statistically significant increase in Cho/Cr ratio from low-grade to HGGs. According to the study by Li et al.,[20] 60% (3 of 5) of Grade III tumors and 81.8% (9 of 11) of Grade IV tumors had elevated lipid/lactate. In our study, we could not effectively differentiate between PCNSL, and HGGs based on the spectroscopy patterns, but the higher Cho/Cr ratio helped in identifying more aggressive lesions. According to Harting et al.,[21] in a study to evaluate the preoperative differential diagnosis of PCNSL and glioma among seven patients with intracranial CNS lymphoma and 21 patients with glioma using single-voxel spectroscopy, all CNS lymphomas demonstrated massively elevated lipid peaks and markedly elevated choline. Similarly, increased lipid peaks were found in only seven necrotic glioblastomas. This allowed them to differentiate PCNSL from solid astrocytomas by massively elevated lipid resonances. They also found higher Cho/Cr ratios in PCNSL than all grades of astrocytomas.
A limitation of our study is the small sample size and fewer cases of PCNSL than HGGs, which make direct comparison difficult. Another limitation is the lack of immunohistochemistry correlation which could have added value to the histopathological diagnosis. Immunohistochemistry markers are useful for prognostication, prediction of treatment response, and patient survival. Advanced MRI techniques can help in preoperative prediction of IDH status, but this could not be effectively confirmed due to the nonavailability of immunohistochemistry data in our patients. We limited our analyses of parameters to the solid, enhancing portion of the lesions. However, an analysis of the peritumoral edema could also add valuable information and could aid in the differential diagnosis. The role of artificial intelligence imaging in diagnosis and evaluation of brain tumors is an exciting field and has a lot of potential in aiding diagnosis.[21],[22]
| Conclusion | | |
The combined use of DWI, PWI, SWI, and spectroscopy is valuable in achieving high sensitivity and specificity in preoperative diagnosis. The DWI was found to be effective in our study in the differentiation of HGGs and PCNSLs with a P = 0.04. The mean ADC in PCNSLs was lower than HGGs. The PWI was found to be effective in their differentiation with a P = 0.001. ITSS was also found to be effective in differentiating HGGs and PCNSLs with a P = 0.001. There was no identifiable unique metabolite on spectroscopy. Multiparametric evaluation of ADC, ITSS, and rCBVmean permits the differentiation of PCNSLs and solid enhancing WHO Grade IV gliomas, which supports the incorporation of advanced MR imaging techniques including DSC-PWI, DWI, and SWI for the routine diagnostic workup of these tumors.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
| References | | |
| 1. | Dasgupta A, Gupta T, Jalali R. Indian data on central nervous tumors: A summary of published work. South Asian J Cancer 2016;5:147-53.  [ PUBMED] [Full text] |
| 2. | Kickingereder P, Wiestler B, Sahm F, Heiland S, Roethke M, Schlemmer HP, et al. Primary central nervous system lymphoma and atypical glioblastoma: Multiparametric differentiation by using diffusion-, perfusion-, and susceptibility-weighted MR imaging. Radiology 2014;272:843-50. |
| 3. | Wang S, Kim S, Chawla S, Wolf RL, Knipp DE, Vossough A, et al. Differentiation between glioblastomas, solitary brain metastases, and primary cerebral lymphomas using diffusion tensor and dynamic susceptibility contrast-enhanced MR imaging. AJNR Am J Neuroradiol 2011;32:507-14. |
| 4. | Doskaliyev A, Yamasaki F, Ohtaki M, Kajiwara Y, Takeshima Y, Watanabe Y, et al. Lymphomas and glioblastomas: Differences in the apparent diffusion coefficient evaluated with high b-value diffusion-weighted magnetic resonance imaging at 3T. Eur J Radiol 2012;81:339-44. |
| 5. | Yamashita K, Yoshiura T, Hiwatashi A, Togao O, Yoshimoto K, Suzuki SO, et al. Differentiating primary CNS lymphoma from glioblastoma multiforme: Assessment using arterial spin labeling, diffusion-weighted imaging, and 18F-fluorodeoxyglucose positron emission tomography. Neuroradiology 2013;55:135-43. |
| 6. | Neska-Matuszewska M, Bladowska J, Sąsiadek M, Zimny A. Differentiation of glioblastoma multiforme, metastases and primary central nervous system lymphomas using multiparametric perfusion and diffusion MR imaging of a tumor core and a peritumoral zone-searching for a practical approach. PLoS One 2018;13:e0191341. |
| 7. | Kayed MH, Saleh TR, Reda IS, Elsirafy MN, Farhoud AH, Abdelzaher E. The added value of advanced neuro-imaging (MR diffusion, perfusion and proton spectroscopy) in diagnosis of primary CNS lymphoma. Alex J Med 2014;50:303-10. |
| 8. | Fawzy FM, Almassry HN, Ismail AM. Preoperative glioma grading by MR diffusion and MR spectroscopic imaging. Egypt J Radiol Nucl Med 2016;47:1539-48. |
| 9. | Boonzaier NR, Larkin TJ, Matys T, van der Hoorn A, Yan JL, Price SJ. Multiparametric MR imaging of diffusion and perfusion in contrast-enhancing and nonenhancing components in patients with glioblastoma. Radiology 2017;284:180-90. |
| 10. | Makino K, Hirai T, Nakamura H, Kuroda JI, Shinojima N, Uetani H, et al. Differentiating between primary central nervous system lymphomas and glioblastomas: Combined use of perfusion-weighted and diffusion-weighted magnetic resonance imaging. World Neurosurg 2018;112:e1-6. |
| 11. | Toh CH, Wei KC, Chang CN, Ng SH, Wong HF. Differentiation of primary central nervous system lymphomas and glioblastomas: Comparisons of diagnostic performance of dynamic susceptibility contrast-enhanced perfusion MR imaging without and with contrast-leakage correction. AJNR Am J Neuroradiol 2013;34:1145-9. |
| 12. | Schob S, Meyer J, Gawlitza M, Frydrychowicz C, Müller W, Preuss M, et al. Diffusion-weighted MRI reflects proliferative activity in primary CNS lymphoma. PLoS One 2016;11:e0161386. |
| 13. | Mangla R, Kolar B, Zhu T, Zhong J, Almast J, Ekholm S. Percentage signal recovery derived from MR dynamic susceptibility contrast imaging is useful to differentiate common enhancing malignant lesions of the brain. AJNR Am J Neuroradiol 2011;32:1004-10. |
| 14. | Hardee ME, Zagzag D. Mechanisms of glioma-associated neovascularization. Am J Pathol 2012;181:1126-41. |
| 15. | Ding Y, Xing Z, Liu B, Lin X, Cao D. Differentiation of primary central nervous system lymphoma from high-grade glioma and brain metastases using susceptibility-weighted imaging. Brain Behav 2014;4:841-9. |
| 16. | Radbruch A, Wiestler B, Kramp L, Lutz K, Bäumer P, Weiler M, et al. Differentiation of glioblastoma and primary CNS lymphomas using susceptibility weighted imaging. Eur J Radiol 2013;82:552-6. |
| 17. | Horská A, Barker PB. Imaging of brain tumors: MR spectroscopy and metabolic imaging. Neuroimaging Clin N Am 2010;20:293-310. |
| 18. | Hourani R, Horská A, Albayram S, Brant LJ, Melhem E, Cohen KJ, et al. Proton magnetic resonance spectroscopic imaging to differentiate between nonneoplastic lesions and brain tumors in children. J Magn Reson Imaging 2006;23:99-107. |
| 19. | Howe 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 Reson Med 2003;49:223-32. |
| 20. | Li X, Zhu Y, Kang H, Zhang Y, Liang H, Wang S, et al. Glioma grading by microvascular permeability parameters derived from dynamic contrast-enhanced MRI and Intratumoral susceptibility signal on susceptibility weighted imaging. Cancer Imaging 2015;15:4. |
| 21. | Harting I, Hartmann M, Jost G, Sommer C, Ahmadi R, Heiland S, et al. Differentiating primary central nervous system lymphoma from glioma in humans using localised proton magnetic resonance spectroscopy. Neurosci Lett 2003;342:163-6. |
| 22. | Rudie JD, Rauschecker AM, Bryan RN, Davatzikos C, Mohan S. Emerging applications of artificial intelligence in neuro-oncology. Radiology 2019;290:607-18. |
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8]
[Table 1], [Table 2]
|
Search Pubmed for