Neurol India Home 
 

ORIGINAL ARTICLE
Year : 2011  |  Volume : 59  |  Issue : 5  |  Page : 674--678

Diffusion magnetic resonance imaging for enhanced visualization of malignant cerebral tumors and abscesses

Shigeo Ohba1, Takakazu Ushioda2, Toru Nakagawa1, Kazuhiko Shimizu3, Hideki Murakami1,  
1 Department of Neurosurgery, Ashikaga Red Cross Hospital, Ashikaga-City, Tochigi, Japan
2 Department of Radiology, Ashikaga Red Cross Hospital, Ashikaga-City, Tochigi, Japan
3 Department of Pathology, Ashikaga Red Cross Hospital, Ashikaga-City, Tochigi, Japan

Correspondence Address:
Shigeo Ohba
Department of Neurosurgery, Ashikaga Red Cross Hospital, 3-2100 Honjo, Ashikaga-City, Tochigi 326-0808
Japan

Abstract

Background: Several malignant tumors and brain abscesses have similar magnetic resonance imaging (MRI) findings. Because the prognosis and treatment strategies differ among these disorders, accurate preoperative diagnosis is very significant. Recently, diffusion-weighted imaging (DWI) and calculation of apparent diffusion coefficients (ADCs) have been reported to be useful to differentiate between the two disorders. Materials and Methods: MRI and DWI were performed in 37 patients: 13 with glioblastomas, 14 with metastatic tumors, 7 with malignant lymphomas, and 3 with brain abscesses. The minimum ADC value (ADC min ) and mean ADC value (ADC mean ) were taken into consideration. Results: Statistically significant differences were found in the enhanced lesions for ADC min and ADC mean between metastatic tumors and malignant lymphomas. The receiver operating characteristic analyses revealed no statistically significant differences among them. There were statistically significant differences in ADC mean in the cores of cystic lesions between brain abscesses and metastatic tumors or glioblastomas. Conclusions: Our results suggest that evaluation using ADC values may facilitate differentiating metastatic tumors from malignant lymphomas, and also metastatic tumors or glioblastomas from brain abscesses.



How to cite this article:
Ohba S, Ushioda T, Nakagawa T, Shimizu K, Murakami H. Diffusion magnetic resonance imaging for enhanced visualization of malignant cerebral tumors and abscesses.Neurol India 2011;59:674-678


How to cite this URL:
Ohba S, Ushioda T, Nakagawa T, Shimizu K, Murakami H. Diffusion magnetic resonance imaging for enhanced visualization of malignant cerebral tumors and abscesses. Neurol India [serial online] 2011 [cited 2020 Jul 2 ];59:674-678
Available from: http://www.neurologyindia.com/text.asp?2011/59/5/674/86539


Full Text

 Introduction



Several malignant tumors such as glioblastomas, metastatic tumors, and malignant lymphomas can have similar features on conventional magnetic resonance imaging (MRI), and brain abscesses are likely to show similar imaging findings as well. Diffusion-weighted imaging (DWI) provides image contrast which is dependent on the molecular motion of water (Brownian motion). [1] DWI is well known to be sensitive for detecting acute cerebral infarction, [2] and is reportedly useful for distinguishing epidermoid from other extra-axial cysts. [3],[4] Recently, DWI and calculation of apparent diffusion coefficients (ADCs) have been studied for their usefulness in differentiating among brain tumors and brain abscesses. [1],[5],[6],[7],[8],[9],[10],[11] In this study, we measured the ADC values in the enhanced solid lesions of malignant cerebral tumors (glioblastomas, metastatic tumors, and malignant lymphomas) and in the cores of the cystic lesions of glioblastomas, metastatic tumors, and brain abscesses, and examined whether these ADC values were useful for distinguishing among the lesions. Moreover, we attempted to determine which of these markers was most useful for distinguishing among the lesions examined.

 Materials and Methods



We studied 37 patients, 23 men and 14 women, who underwent surgical procedures between 2004 and 2010 at the Ashikaga Red Cross Hospital, Ashikaga-city, Japan. Patients' age ranged between 35 and 83 years. The lesions included: glioblastomas (13), metastatic tumors (14), malignant lymphomas (7), and cerebral abscesses (3). In patients with metastatic tumors, the primary was: lung carcinoma in 7, breast cancer in 2, and endometrial cancer, ovarian cancer, bladder cancer, thymoma, and tumor of unknown origin one each. None of the patients had undergone chemotherapy or radiotherapy before the MRI examination. The diagnosis was based on operative specimens.

Conventional MRI and DWI were performed using the same instrument, and images were retrospectively evaluated. The imaging was performed on a 1.0 T system (Magnetom Harmony; Siemens, Erlangen, Germany). The MRI protocol consisted of T1-weighted spin-echo (SE) images (TR/TE: 615/12 ms), T2-weighted turbo SE (TR/TE: 4 000/98 ms), and fluid-attenuated inversion-recovery sequences (TR/TE: 9 000/119 ms). DWI was obtained in the axial plane using an echo-planar sequence with the following parameters: TR/TE 4 500/83 ms; matrix size, 256× 256; FOV, 220 mm; slice thickness, 5 mm; and diffusion gradient encoding in three orthogonal directions (x, y, and z axes) at b values of 0 and 1000 s/mm 2 . ADC map images were generated automatically by the MRI unit. The DWI was analyzed by placing regions of interest (ROI) over the tumor on ADC map images. The ROI placement and ADC calculations were made from the enhanced solid portions of the tumor and cores of the cystic portions, respectively. At least three measurements were made and the minimum ADC value (ADC min ) and mean ADC value (ADC mean ) were taken into consideration.

ANOVA was used to calculate whether or not these lesions showed significant differences according to the different ADC values and ratios. For further investigation of the significant differences in ADC values, Scheffe's F test was used. Receiver operating characteristic (ROC) analysis was performed using the ROCkit toolkit (http://xray.bsd.uchicago.edu/krl/roc_soft.htm). A P value less than 0.05 was considered to indicate a statistically significant difference.

 Results



The findings of T1-weighted imaging with gadolinium diethylenetriaminepentaacetic acid (DTPA) and DWI are summarized in [Table 1]. Approximately 60% of glioblastomas and 40% of metastatic tumors showed ring enhancement with gadolinium DTPA. None of the malignant lymphomas but all the three brain abscesses showed ring enhancement. Approximately one-third of metastatic tumors, half of glioblastomas, most of malignant lymphomas, and all brain abscesses showed high intensity on DWI.{Table 1}

As to ADC, we first determined the ADC min and ADC mean in the enhanced regions of glioblastomas, metastatic tumors, and malignant lymphomas. There were statistically significant differences in ADC mean and ADC min in the enhanced lesions between metastatic tumors and malignant lymphomas, but not between metastatic tumors and glioblastomas, or between glioblastomas and malignant lymphomas [Table 2]. To determine which is a more useful predictor, ROC analyses for ADC min and ADC mean between metastatic tumors and malignant lymphomas were performed. The areas under the curve of ADC min and ADC mean were 0.88 and 0.85, respectively. There were no statistically significant differences between them. Our results showed ADC min and ADC mean to be similarly useful for distinguishing metastatic tumors from malignant lymphomas.{Table 2}

Furthermore, we measured the ADC mean in the cores of cystic regions of glioblastomas, metastatic tumors, and cerebral abscesses. There was a statistically significant difference in ADC mean in the cores of cystic lesions between metastatic tumors and abscesses, and between glioblastomas and abscesses, but not between metastatic tumors and glioblastomas [Table 3].{Table 3}

 Discussion



In the studies focusing on the usefulness of ADC values for distinguishing cerebral neoplasms from brain abscesses, some authors have used ADC mean and others have used ADC min . [1],[5],[6],[7],[8],[9],[10],[11] The ADC values of cerebral tumors reportedly correlated with tumor cellularity. [7],[12] The reason for selecting ADC min was reported to be sampling the most cellular area because there are regions of variable cellularity within a tumor. [6] The data from several previous studies as well as ours are summarized in [Table 4] and [Table 5]. The ADC values in our study were found to be similar to those in other studies.{Table 4}{Table 5}

We first focused on the solid portions of malignant tumors. The differentiation of metastatic tumors from other malignant tumors is relatively simple, if the primary cancer has been identified and there are multiple lesions in the brain. However, it is very difficult to diagnose metastatic tumors if no primary cancer is detected and a solitary tumor is found in the brain. High intensity on DWI is among the characteristic findings of malignant lymphomas, [13] but glioblastomas and metastatic tumors occasionally show high intensity on DWI. Therefore, the finding of high intensity on DWI alone is not very useful for distinguishing malignant lymphoma from glioblastomas and metastatic tumors. More accurate preoperative diagnosis of malignant lymphoma is highly significant, because it is believed to be meaningless to resect malignant lymphomas aggressively.

Our data showed ADC min and ADC mean to differ significantly between metastatic tumors and malignant lymphomas. Yamasaki et al. found the ADC mean of malignant lymphoma s to be lower than that of metastatic tumors, [11] which is consistent with our results. On the contrary, Calli et al. reported that there was no difference in ADC min between malignant lymphoma s and metastatic tumors. [5] Kitis et al. reported that the ADC min of malignant lymphomas seemed to be lower than that of metastatic tumors, though the difference did not reach statistical significance. [6] To evaluate which is the better predictive marker among these values for distinguishing metastatic tumors from malignant lymphomas, ROC analyses were performed between ADC min and ADC mean . These analyses revealed no statistically significant differences between them. Our results suggest ADC min and ADC mean to be of similar value as predictive factors for distinguishing metastatic tumors from malignant lymphomas.

ADC min and ADC mean of enhanced regions were found not to differ significantly between metastatic tumors and glioblastomas in our study. Several published studies obtained the same results, [5],[6],[7],[11] though Krabbe et al. reported the ADC mean of metastatic tumors to be significantly higher than that of malignant gliomas (glioblastoma, anaplastic astrocytoma, and anaplastic oligodendroglioma). [8] On the contrary, Server et al. reported ADC min and ADC mean of metastatic tumors to be lower than those of high-grade gliomas. [10]

There was no statistically significant difference in ADC min or ADC mean between malignant lymphomas and glioblastomas in our study, although the ADC values of glioblastomas appeared to be higher than those of malignant lymphomas. Yamasaki et al. found the average ADC of malignant lymphomas to be lower than that of glioblastomas. [11] Kitis et al. described lymphomas as seeming to have markedly restricted diffusion coefficients. [6] As the numbers of such lesions were small in our study, further examination of more cases may reveal statistically significant differences in ADC values between these lesions.

Furthermore, we focused on the cores of cystic lesions. MR imaging of glioblastomas, metastatic tumors, and brain abscesses often shows ring enhancement with gadolinium DTPA. Although the characteristic features of brain abscesses are of high intensity on DWI, glioblastomas or metastatic tumors occasionally show high intensity on DWI. Several reports have described brain abscesses as having markedly low signals in the abscess capsule on T2-weighted images and that the capsule on the ventricular side was thinner than that on the cortical side. [14],[15] However, these features are not consistently seen. Therefore, accurately distinguishing among these lesions is not easy. Because the treatments and prognosis of brain abscesses are far different from those of malignant tumors, it is very important to preoperatively distinguish an abscess from a glioblastoma or metastatic tumor. Moreover, it is also important to distinguish a brain abscess from tumor recurrence in the surgical bed after removal of brain tumors.

There were statistically significant differences in ADC mean between brain abscesses and glioblastomas or metastatic tumors, but not between glioblastomas and metastatic tumors. These results were consistent with those of other reports. Chang et al. and Fertikh et al. concluded that DWI is more useful than conventional MR imaging for distinguishing brain abscesses from cystic or necrotic tumors. [16],[17] Abscess pus is reportedly a creamy viscous fluid containing inflammatory cells, bacteria, mucoid proteins, and cell debris. The viable cell density in pus is the main biological parameter responsible for decreased diffusion in brain abscesses. [18] We should keep in mind that the age of an abscess and treatment with antibiotics might alter the ADC values. Some brain abscesses, such as those due to toxoplasmosis, reportedly show high ADC values. [16],[19] This was reported to be due to the core of a toxoplasma abscess consisting primarily of necrotic tissue and lacking the viscous, proteinaceous, inflammatory debris of purulent fluid. [19] The contents of glioblastomas and metastatic tumors are considered to be partially necrotic, thereby producing high ADC values, though some authors have reported low ADC values in the necrotic part of the tumor. [20],[21] Holtas et al. reported the low ADC value to be due to early necrosis with intracellular edema of the lesions. When diagnosing these lesions, we need to keep such atypical cases in mind. [21]

MR spectroscopy was reported to be effective for differentiating brain abscesses from cystic brain tumors. However, there are several limitations to MR spectroscopy which include longer acquisition times, limited coverage, volume averaging with normal brain tissue in the case of small masses, and greater susceptibility to field inhomogeneity near the skin. Therefore, DWI is considered to be more effective, especially for small lesions near the skull. [17]

In conclusion our results suggest that evaluation using ADC values might facilitate differentiating between metastatic tumors and malignant lymphomas, and for distinguishing metastatic tumors or glioblastomas from cerebral abscesses.

References

1Dorenbeck U, Butz B, Schlaier J, Bretschneider T, Schuierer G, Feuerbach S. Diffusion-weighted echo-planar MRI of the brain with calculated ADCs: A useful tool in the differential diagnosis of tumor necrosis from abscess? J Neuroimaging 2003;13:330-8.
2Lövblad KO, Laubach HJ, Baird AE, Curtin F, Schlaug G, Edelman RR, et al. Clinical experience with diffusion-weighted MR in patients with acute stroke. AJNR Am J Neuroradiol 1998;19:1061-6.
3Laing AD, Mitchell PJ, Wallace D. Diffusion-weighted magnetic resonance imaging of intracranial epidermoid tumours. Australas Radiol 1999;43:16-9.
4Tsuruda JS, Chew WM, Moseley ME, Norman D. Diffusion-weighted MR imaging of the brain: Value of differentiating between extraaxial cysts and epidermoid tumors. AJR Am J Roentgenol 1990;155:1059-65.
5Calli C, Kitis O, Yunten N, Yurtseven T, Islekel S, Akalin T. Perfusion and diffusion MR imaging in enhancing malignant cerebral tumors. Eur J Radiol 2006;58:394-403.
6Kitis O, Altay H, Calli C, Yunten N, Akalin T, Yurtseven T. Minimum apparent diffusion coefficients in the evaluation of brain tumors. Eur J Radiol 2005;55:393-400.
7Kono K, Inoue Y, Nakayama K, Shakudo M, Morino M, Ohata K, et al. The role of diffusion-weighted imaging in patients with brain tumors. AJNR Am J Neuroradiol 2001;22:1081-8.
8Krabbe K, Gideon P, Wagn P, Hansen U, Thomsen C, Madsen F. MR diffusion imaging of human intracranial tumours. Neuroradiology 1997;39:483-9.
9Muccio CF, Esposito G, Bartolini A, Cerase A. Cerebral abscesses and necrotic cerebral tumours: Differential diagnosis by perfusion-weighted magnetic resonance imaging. Radiol Med 2008;113:747-57.
10Server A, Kulle B, Maehlen J, Josefsen R, Schellhorn T, Kumar T, et al. Quantitative apparent diffusion coefficients in the characterization of brain tumors and associated peritumoral edema. Acta Radiol 2009;50:682-9.
11Yamasaki F, Kurisu K, Satoh K, Arita K, Sugiyama K, Ohtaki M, et al. Apparent diffusion coefficient of human brain tumors at MR imaging. Radiology 2005;235:985-91.
12Sugahara T, Korogi Y, Kochi M, Ikushima I, Shigematu Y, Hirai T, et al. Usefulness of diffusion-weighted MRI with echo-planar technique in the evaluation of cellularity in gliomas. J Magn Reson Imaging 1999;9:53-60.
13Akter M, Hirai T, Makino K, Kitajima M, Murakami R, Fukuoka H, et al. Diffusion-weighted imaging of primary brain lymphomas: Effect of ADC value and signal intensity of T2-weighted imaging. Comput Med Imaging Graph 2008;32:539-43.
14Falcone S, Post MJ. Encephalitis, cerebritis, and brain abscess: Pathophysiology and imaging findings. Neuroimaging Clin N Am 2000;10:333-53.
15Haimes AB, Zimmerman RD, Morgello S, Weingarten K, Becker RD, Jennis R, et al. MR imaging of brain abscesses. AJR Am J Roentgenol 1989;152:1073-85.
16Chang SC, Lai PH, Chen WL, Weng HH, Ho JT, Wang JS, et al. Diffusion-weighted MRI features of brain abscess and cystic or necrotic brain tumors: Comparison with conventional MRI. Clin Imaging 2002;26:227-36.
17Fertikh D, Krejza J, Cunqueiro A, Danish S, Alokaili R, Melhem ER. Discrimination of capsular stage brain abscesses from necrotic or cystic neoplasms using diffusion-weighted magnetic resonance imaging. J Neurosurg 2007;106:76-81.
18Mishra AM, Gupta RK, Saksena S, Prasad KN, Pandey CM, Rathore D, et al. Biological correlates of diffusivity in brain abscess. Magn Reson Med 2005;54:878-85.
19Chong-Han CH, Cortez SC, Tung GA. Diffusion-weighted MRI of cerebral toxoplasma abscess. AJR Am J Roentgenol 2003;181:1711-4.
20Hakyemez B, Erdogan C, Yildirim N, Parlak M. Glioblastoma multiforme with atypical diffusion-weighted MR findings. Br J Radiol 2005;78:989-92.