Diagnostic Value of Peritumoral Minimum Apparent Diffusion Coefficient for Differentiation of Glioblastoma Multiforme From Solitary Metastatic Lesions

Neuroradiology/Head and Neck Imaging Original Research Lee et al. MRI to Diagnose Glioblastoma Multiforme Neuroradiology/Head and Neck Imaging Original Research FOCUS ON: Eun Ja Lee 1,2 Karel terbrugge 2 David Mikulis 2 Dae Seob Choi 2,3 Jong Myon Bae 4 Seon Kyu Lee 5 Soon Young Moon 1 Lee EJ, terbrugge K, Mikulis D, et al. Keywords: brain imaging, brain metastases, diffusionweighted imaging, glioblastoma multiforme, MRI DOI:10.2214/AJR.10.4752 Received April 7, 2010; accepted after revision May 13, 2010. 1 Department of Radiology, Kwandong University, College of Medicine, Myongji Hospital, Koyang City, Gyunggi-Do, South Korea. 2 Department of Medical Imaging, Division of Neuroradiology, Toronto Western Hospital, 399 Bathurst St., 3MCL-434, Toronto, ON M5T2S8, Canada. Address correspondence to K. terbrugge (karel.terbrugge@uhn.on.ca). 3 Department of Radiology, Gyeongsang National University, College of Medicine, Jinju, South Korea. 4 Department of Preventive Medicine, Jeju National University, Jeju, South Korea. 5 Department of Radiology, Lahey Clinic Medical Center, Burlington, MA. AJR 2011; 196:71 76 0361 803X/11/1961 71 American Roentgen Ray Society Diagnostic Value of Peritumoral Minimum Apparent Diffusion Coefficient for Differentiation of Glioblastoma Multiforme From Solitary Metastatic Lesions OBJECTIVE. In glioblastoma multiforme, the peritumoral region may be infiltrated with malignant cells in addition to vasogenic edema, whereas in a metastatic deposit, the peritumoral areas comprise predominantly vasogenic edema. The purpose of this study was to determine whether the minimum apparent diffusion coefficient (ADC) can be used to differentiate glioblastoma from solitary metastasis on the basis of cellularity levels in the enhancing tumor and in the peritumoral region. MATERIALS AND METHODS. Seventy-three patients underwent conventional MRI and diffusion-weighted imaging (DWI) before undergoing treatment. The minimum ADC was measured in the enhancing tumor, peritumoral region, and contralateral normal white matter. To determine whether there was a statistical difference between metastasis and glioblastoma, we analyzed patient age and sex, minimum ADC value, and ADC ratio of the two groups. A receiver operating characteristic (ROC) curve analysis was used to determine the cutoff value of the minimum ADC that had the best combination of sensitivity and specificity for distinguishing between glioblastoma and metastasis. RESULTS. The mean minimum ADC values and mean ADC ratios in the peritumoral regions of glioblastomas were significantly higher than those in metastases. However, the mean minimum ADC values and mean ADC ratios in enhancing tumors showed no statistically significant difference between the two groups. According to ROC curve analysis, a cutoff value of 1.302 10 3 mm 2 /s for the minimum peritumoral ADC value generated the best combination of sensitivity (82.9%) and specificity (78.9%) for distinguishing between glioblastoma and metastasis. CONCLUSION. Although the characteristics of solitary metastasis and glioblastoma multiforme may be similar on conventional MRI, DWI can offer diagnostic information to distinguish between the tumors. T he two most common malignant brain neoplasms are glioma and metastasis. In many cases, differentiation of the two neoplasms is suggested from the clinical history or the presence of lesions elsewhere. In many clinical settings, especially in patients with multiple lesions, the diagnosis of brain metastasis is usually straightforward and uncomplicated. However, this distinction may be difficult when patients present with a solitary enhancing lesion. On conventional MR images, highgrade glioma and solitary metastatic brain tumor often display similar signal intensity characteristics and contrast enhancement patterns [1, 2]. In many cases, a biopsy is performed for histologic confirmation even if there is a history of a known primary malignancy [3 6]. In peritumoral edema of primary highgrade glioma, infiltrating neoplastic cells have been reported, whereas in metastasis, peritumoral edema consists essentially of vasogenic edema. Therefore, the key to distinguishing between these two entities appears to lie in detecting the changes within the peritumoral area that is, the area beyond the enhancing margin on imaging. Many new developments in MRI techniques have been used in efforts to make this distinction in the peritumoral area, including spectroscopy, perfusion imaging, diffusion tensor imaging, and the measurement of the absolute apparent diffusion coefficient (ADC) [3 5]. The introduction of diffusion-weighted imaging (DWI) has enabled us to obtain additional information about the brain from AJR:196, January 2011 71

Lee et al. the microscopic movement of water protons. DWI has been used to grade or differentiate among brain tumors on the basis of cellularity [7]. We hypothesized that the ADC values within an enhancing tumor and within the peritumoral area provide quantitative information on tumor cellularity and characterization of tumor-related edema that is not readily discernible on conventional MRI and that these changes can be used to differentiate glioblastomas from metastatic lesions. Few studies have evaluated the usefulness of the minimum ADC value in the tumoral and peritumoral regions to differentiate glioblastoma from solitary metastasis in a large study population. The purpose of this study was to determine whether a minimum ADC value can be used to differentiate glioblastomas from solitary metastases on the basis of cellularity levels in the contrast-enhancing tumors and nonenhancing peritumoral edema regions. Materials and Methods This study was reviewed and approved by the institutional review board. Patients The MRI examinations of 73 consecutive patients (41 men, 32 women; age range, 29 83 years; mean age, 57.5 ± 12.3 years) with a diagnosis of glioblastoma multiforme or solitary metastasis were evaluated retrospectively. All patients had a previously untreated solitary enhancing brain tumor and peritumoral edema and had undergone conventional brain MRI and DWI before surgical intervention at our institution between January 2004 and June 2006. In each case, a histopathologic diagnosis based on World Health Organization (WHO) criteria was determined using surgical specimens. In all patients, tumor diagnosis was verified histologically at surgical resection. Stereotactic biopsy was not performed in any of the patients. Of the 73 patients, WHO grade IV glioblastoma was diagnosed in 38 cases and solitary metastasis in 35. Metastatic brain tumors included lung carcinoma (n = 16), breast carcinoma (n = 3), anorectal carcinoma (n = 3), melanoma (n = 3), renal carcinoma (n = 2), colon carcinoma (n = 2), thyroid carcinoma (n = 1), sarcoma of the thigh (n = 1), esophageal carcinoma (n = 1), and carcinoma of unknown origin (n = 3). Two of the authors reviewed the MRI and DWI findings in tandem. MRI and Image Processing All MRI examinations were performed on a 1.5-T system (Signa EchoSpeed, version 8.2.3 software, GE Healthcare) with a standard head coil. Conventional MR images, including FLAIR images (TR/TE, 9,000/165; inversion time, 2,200 milliseconds; number of signals acquired, 1; section thickness, 5 mm; intersection gap, 2 mm; matrix, 256 192; field of view [FOV], 22.0 22.0 cm), T1-weighted images (450/20; number of signals acquired, 1; section thickness, 5 mm; intersection gap, 2 mm; matrix, 256 224; FOV, 21.9 21.9 cm), T2-weighted images (4,250/93; number of signals acquired, 1; section thickness, 5 mm; intersection gap, 2 mm; matrix, 256 256; FOV, 21.9 21.9 cm), and contrast-enhanced T1-weighted images, and DW images were obtained during the same examination. DW images were acquired in the transverse plane using a spin-echo echo-planar sequence, with the diffusion gradient encoded in three orthogonal directions (11,000/59; number of signals acquired, 1; section thickness, 5 mm; intersection gap, 0; matrix, 128 128; FOV, 29.9 29.9 cm) with three orthogonal directional motionprobing gradients (b = 1,000 s/mm 2 ), followed by the automatic generation of isotropic DWI. Images without motion-probing gradients (b = 0 s/mm 2 ) were obtained simultaneously. ADC maps were calculated from isotropic DWI, and images with a b value of 0 s/mm 2 were obtained. For the MR examinations of 18 patients, exponential DWI was performed. Eleven of these patients had metastasis and seven had glioblastoma. The exponential images provided a more accurate depiction of diffusion effects than trace DWI by removing the contribution from T2 signal intensity of the tissue being examined [8]. We also evaluated susceptibility artifact on T2* gradient-echo images to assess for hemorrhage in the enhancing tumors because intratumoral hemorrhage or susceptibility artifact may affect ADC values. Sixty-two patients underwent T2* gradient-echo imaging: 30 patients had metastasis and 32 had glioblastoma. In all patients, MRI was performed the same day as surgery. Normal white matter, enhancing tumoral area, and the peritumoral region were defined on the basis of the following imaging features: normal white matter, normal-appearing mirrored areas that contained no enhancement and showed normal signal intensity on T2 imaging, FLAIR imaging, and DWI; enhancing tumoral area, a region that contained a solid portion, preferably with avoidance of cystic or necrotic components; and peritumoral area, a region clearly outside the welldefined enhancing solid portion that contained absolutely no enhancement and had high signal intensity on T2-weighted imaging. The minimum ADC of each tumor was determined by placing regions of interest (ROIs) using a workstation (Advantage, GE Healthcare) for Microsoft Windows operating with the FuncTool software program (GE Healthcare); ROI placement was performed by one neuroradiologist without knowledge of the histologic information. First, we selected all continuous sections that included enhancing tumor and the peritumoral region. For exact determination of tumoral heterogeneity, at least three or four uniform round or oval ROIs (area ~ 40 mm 2 ) were placed carefully on each selected section of the ADC map including the areas of enhancing tumor and peritumoral edema with the lowest ADC determined by visual inspection. The ROIs were positioned carefully to avoid contamination from different adjacent tissues with reference to conventional MRI. Finally, the ROI with the lowest ADC was chosen from these ROIs as the minimum ADC. The peritumoral ADC ratio was calculated by dividing the minimum ADC value of the peritumoral edema of the affected hemisphere by that of the normal white matter of the contralateral hemisphere. The tumoral ADC ratio was calculated by dividing the ADC value of the enhancing tumor of the affected hemisphere by that of the normal white matter of the contralateral hemisphere. Diffusion changes caused by the administration of a steroid or other agents could be neglected in this series. Statistical Analysis In all patients, the minimum ADC value ( 10 9 mm 2 /s) from DWI was measured in the enhancing tumor, peritumoral region, and contralateral normal white matter. To determine whether there was a the statistical difference between metastasis and glioblastoma, we analyzed patient age and sex, minimum ADC value, and minimum ADC ratio in the two groups using the Wilcoxon rank sum test or chi-square test. A receiver operating characteristic (ROC) analysis was used to determine the cutoff value of the minimum ADC that had the best combination of sensitivity and specificity for distinguishing between glioblastoma and metastasis. A p value of < 0.05 indicated a statistically significant difference. Results Patient age and sex, minimum ADC value, and minimum ADC ratio in the glioblastoma and metastasis groups are shown in Table 1. The mean minimal ADC value in the peritumoral region of glioblastomas (1.149 ± 0.119 [SD]) was significantly lower than that in metastases (1.413 ± 0.147) (p < 0.05). The mean peritumoral ADC ratio was also significantly lower in glioblastomas (1.466 ± 0.24) than in metastases (1.829 ± 0.25) (p < 0.05) 72 AJR:196, January 2011

MRI to Diagnose Glioblastoma Multiforme TABLE 1: Patient Age and Sex, Mean Minimum Apparent Diffusion Coefficient (ADC) Values, and Mean ADC Ratios for Glioblastomas and Metastases Patient Characteristic or MRI Finding (Table 1 and Figs. 1 3). The mean minimum tumoral ADC values (glioblastoma, 902.8 ± 226.54; metastasis, 893.6 ± 216.10) and tumoral ADC ratios (glioblastoma, 1.152 ± 0.29; metastasis, 1.145 ± 0.25) showed no statistically significant difference between the two groups (p > 0.05) (Table 1 and Figs. 1 3). There was a statistically significant sex difference between the two groups. However, the statistical results between the two tumor groups were unchanged even after stratifying sex. There was no statistically significant age difference between the two patient groups. According to the ROC curve analysis, a cutoff value of 1.302 10 3 mm 2 /s for the peritumoral minimum ADC generated the best combination of sensitivity (82.9%) Glioblastoma Multiforme (n = 38) Metastasis (n = 35) p a Age (y) 57 ± 13.5 (56.5) 58 ± 11.0 (58.0) 0.73 Sex (% of men) 68.4 42.9 0.03 Minimum ADC value ( 10 3 mm 2 /s) Tumoral region 0.903 ± 0.227 (0.884) 0.894 ± 0.216 (0.897) 0.89 Peritumoral region 1.149 ± 0.119 (1.161) 1.413 ± 0.147 (1.402) < 0.0001 ADC ratio Tumoral region b 1.152 ± 0.29 (1.145) 1.145 ± 0.25 (1.172) 0.91 Peritumoral region c 1.466 ± 0.24 (1.476) 1.829 ± 0.25 (1.860) < 0.0001 Note Except p values and sex data, data are presented as mean ± SD (median). a Wilcoxon rank sum test or chi-square test. b Minimum tumoral ADC / contralateral normal white matter ADC. c Minimum peritumoral ADC / contralateral normal white matter ADC. Fig. 1 Glioblastoma multiforme in right temporal lobe of 54-year-old man. A, Contrast-enhanced T1-weighted image shows inhomogeneous enhancing mass in right temporal lobe with extensive peritumoral edema. B, T2-weighted image shows hyperintense mass lesion associated with extensive peritumoral edema. C, Exponential diffusion-weighted image reflects only diffusional properties and mass has decreased signal intensity. Peritumoral edema reveals different signal intensities. Infiltrative edema (long arrows) shows relatively intermediate signal intensity, and vasogenic edema (short arrows) shows markedly decreased signal intensity. D, On apparent diffusion coefficient (ADC) map, mass has central cystic changes. Peritumoral edema appears inhomogeneously hyperintense relative to normal brain tissue. Infiltrative edema (arrows) has intermediate ADC values. Measured minimum ADC was 1.190 10 3 mm 2 /s within peritumoral region. Measured minimum ADC was 1.149 10 3 mm 2 /s within enhancing mass (not shown). Circles show regions of interest. and specificity (78.9%) (p < 0.05) for distinguishing between glioblastoma and metastasis (Fig. 4). A C On T2* gradient-echo images, in the tumoral core of metastases, 17 of 30 patients (57%) had various susceptibility artifacts (focal, mottled, patchy, and near total), and in the tumoral core of glioblastomas, 24 of 32 patients (75%) had various susceptibility artifacts (focal, mottled, patchy, and near total). Discussion Gliomas and metastases are the most frequent brain tumors. It is clinically important to differentiate glioblastoma from a single brain metastasis because medical staging, surgical planning, and therapeutic decisions are vastly different for each tumor type [5, 6]. Advanced MRI techniques are used increasingly to obtain physiologic and metabolic information that complements the anatomic information provided by conventional MRI [6]. DWI has been used to grade or identify brain tumors on the basis of cellularity. The ADC measured within the brain reflects the mobility of the free water fraction including B D AJR:196, January 2011 73

Lee et al. extracellular and intracellular water within the tissue. Several studies have shown that ADC correlates well with tumor cellularity on histologic examination and that calculation of ADC may aid conventional MRI in characterizing cerebral tumors [7, 9 12]. Previous studies have shown that tumoral ADC is not useful for distinguishing between glioblastomas and metastatic tumors [1, 7, 13 15]. However, Krabbe et al. [10] and Chiang et al. [4] found that tumoral ADC of cerebral metastasis is significantly higher than that of high-grade astrocytoma. In the current study, tumoral ADC showed no statistically significant difference between the two groups. Glioblastomas and metastatic tumors often contain heterogeneous signal intensity secondary to necrosis and susceptibility artifacts. As a result of this heterogeneity, DWI metrics obtained from the tumor can be imprecise or inaccurate. Many A C patients in our study had various susceptibility artifacts on T2* gradient-echo imaging in the tumoral part of the metastasis or glioblastoma. Most brain tumors are surrounded by a T2 high-signal abnormality that traditionally has been termed vasogenic edema. Vasogenic edema is the most frequent form of brain edema associated with brain tumors. Local disruption of the blood brain barrier increases capillary permeability and induces a pressure gradient from the vascular to extracellular compartment that results in the retention of plasma fluid and protein in the extracellular spaces [1]. In general, the nonenhancing area of an abnormality that surrounds the enhancing tumor core is referred to as peritumoral edema. In metastatic brain tumors or noninfiltrative primary tumors such as meningioma, peritumoral edema is synonymous with vasogenic edema, B D Fig. 2 Metastasis from colon adenocarcinoma in left frontal lobe of 59-year-old woman. A, Contrast-enhanced T1-weighted image shows well-defined, uneven peripheral enhancing mass in left frontal lobe. B, Mass has inhomogeneous signal intensity associated with extensive peritumoral edema on T2- weighted image. C, Exponential diffusion-weighted image shows cystic mass lesion with intermediate-signal rim. Peritumoral edema shows relatively homogeneous low signal intensity, which reflects pure vasogenic edema. D, On apparent diffusion coefficient (ADC) map, ADC value is homogeneously elevated in peritumoral region. Enhancing mass reveals reverse signal intensity with exponential diffusion-weighted imaging. Measured minimum ADC was 1.606 10 3 mm 2 /s within peritumoral region. Measured minimum ADC with lowest ADC was 0.808 10 3 mm 2 /s within enhancing mass. Circles show regions of interest. in which increased extracellular water from leakage of plasma fluid due to altered tumor capillaries is present but no tumor cells are present. In glioma, however, peritumoral edema is better referred to as infiltrative edema because it represents vasogenic edema and infiltrating tumor cells that are behind the blood brain barrier and that usually invade along the white matter tracts. Differentiation of vasogenic edema from infiltrative edema has been attempted using DWI on the basis of the premise that water diffusivity is facilitated to a greater degree in vasogenic edema than in infiltrative edema because of a lack of intervening tumor cells in the former [1, 16]. Several studies have shown that peritumoral ADC is useful for distinguishing between glioblastomas and metastatic tumors [4, 10, 14]. It has been hypothesized that ADC values could delineate areas of neoplastic cell infiltration [4, 10, 14]. Investigators have reported that brain neoplasms with higher cellularity show a significant reduction in ADC value [7, 9, 10]. We included only the minimum ADC value measurements. The regions with minimum ADCs have been suggested to reflect the highest tumor cell density, or the most proliferative portion of the tumor, within heterogeneous tumors; therefore, these sites may be of diagnostic value in identifying infiltrative peritumoral edema. Recent studies have shown that minimum ADCs may facilitate accurate grading of astrocytic tumors because regions exhibiting the minimum ADC correspond to the highest-grade glioma foci within heterogeneous tumors [12, 17, 18]. Some studies have shown that the minimum ADC values of tumors have 74 AJR:196, January 2011

MRI to Diagnose Glioblastoma Multiforme 2.00 Glioblastoma Metastasis 2.50 Glioblastoma Metastasis Sensitivity Minimum ( 10 3 mm 2 /s) ADC Value 1.0 0.8 0.6 0.4 0.2 1.75 1.50 * 1.25 1.00 0.75 0.50 0.0 0.0 Peritumoral edema 0.2 0.4 0.6 1 Specificity preoperative prognostic importance in patients with malignant supratentorial astrocytoma [19, 20]. Furthermore, measuring the lowest ADC within a tumor might aid in selecting an appropriate site for a stereotactic biopsy. We found that the minimum ADC value of peritumoral edema in glioblastomas was significantly lower than that in metastases. Enhancing tumor 0.8 1.0 A ADC Ratio 2.25 2.00 * 1.75 1.50 1.25 1.00 0.75 0.50 Peritumoral edema Enhancing tumor Fig. 3 Box-and-whisker plots. Thick horizontal line = mean, whiskers = ± SD. A and B, Minimum apparent diffusion coefficient (ADC) values (A) and ADC ratios (B) in glioblastoma multiforme and metastasis show that minimum peritumoral ADC values and peritumoral ADC ratios in metastasis were significantly higher than those in glioblastoma (p < 0.05). Asterisks represent statistically significant differences. Minimum tumoral ADC values and tumoral ADC ratios showed no significant difference between the two groups (p > 0.05). Fig. 4 Empiric receiver operating characteristic (ROC) curve of minimum peritumoral apparent diffusion coefficient (ADC) values for use in differentiation of glioblastoma multiforme from metastasis. Area under ROC curve was 0.879 (95% CI, 0.80 0.96). When cutoff value of minimum peritumoral ADC was 1.302 10 3 mm 2 /s, best combination of sensitivity (82.9%) and specificity (78.9%) was provided. This finding may help to distinguish preoperatively between glioblastomas and metastases. The higher minimum ADC in the peritumoral regions of metastases suggests there are higher intracellular and extracellular water fractions than in glioblastomas. Our results support the hypothesis that minimum ADC values can detect neoplastic cell infiltration in peritumoral edema in patients with glioblastoma. Therefore, analysis of these peritumoral regions may prove to be more robust than analysis of the lesion itself. However, a few reports do not support the hypothesis that peritumoral neoplastic cell infiltration is depicted by ADC values [9, 21]. Other new developments in MRI techniques have been used to test this distinction in the peritumoral area including spectroscopy, perfusion imaging, and diffusion tensor imaging [3 6, 22]. Several investigators have shown that perfusion-weighted imaging and MR spectroscopy of the peritumoral region can be used to show differences between solitary metastasis and high-grade glioma [4, 6, 22]. Their results have shown that the cholineto-creatine ratio and relative cerebral blood volume in the peritumoral region of highgrade glioma are significantly higher than in metastasis. However, perfusion-weighted and proton spectroscopic MRI in the enhancing portion of the lesion cannot be used to discriminate reliably between metastasis and high-grade glioma [4, 6, 14, 23 25]. Diffusion tensor imaging has been used to distinguish between solitary brain metastasis and high-grade glioma. Using mean diffusivity and anisotropy measures, several investigators have tried to differentiate tumor-infiltrated edema associated with high-grade glioma B AJR:196, January 2011 75

Lee et al. from purely vasogenic edema associated with metastasis. In studies comparing metastasis with high-grade glioma, peritumoral fractional anisotropy was not significantly different, whereas mean diffusivity of the peritumoral edema was higher for metastasis than for high-grade glioma [2, 4, 5, 26]. Of the various advanced noninvasive techniques, we evaluated the usefulness of only minimum ADC in the contrast-enhancing tumoral and peritumoral regions to differentiate glioblastoma from solitary metastasis because DWI should be available in most hospitals and is the easiest and least timeconsuming technique. The processing of the data is also simple. Our study has several limitations. Biopsy of peritumoral edematous areas was not performed for histologic examination at the time of surgery. We used a retrospective approach to select cases and subjective placement of the ROIs. In conclusion, conventional MRI is often unreliable for the differentiation of glioblastoma multiforme from solitary metastatic lesions. Our results showed that peritumoral minimum ADC values can be a powerful tool in the diagnosis and surgical management of glioblastomas and metastatic tumors. We believe that the minimum ADC values in the peritumoral regions can provide valuable information for differentiating glioblastoma multiforme from a solitary metastasis. References 1. Oh J, Cha S, Aiken AH, et al. Quantitative apparent diffusion coefficients and T2 relaxation times in characterizing contrast enhancing brain tumors and regions of peritumoral edema. J Magn Reson Imaging 2005; 21:701 708 2. Lu S, Ahn D, Johnson G, Cha S. Peritumoral diffusion tensor imaging of high-grade gliomas and metastatic brain tumors. AJNR 2003; 24:937 941 3. Tang YM, Ngai S, Stuckey S. The solitary enhancing cerebral lesion: can FLAIR aid the differentiation between glioma and metastasis? AJNR 2006; 27:609 611 4. Chiang IC, Kuo YT, Lu CY, et al. Distinction between high-grade gliomas and solitary metastases using peritumoral 3-T magnetic resonance spectroscopy, diffusion, and perfusion imagings. Neuroradiology 2004; 46:619 627 5. Tsuchiya K, Fujikawa A, Nakajima M, Honya K. Differentiation between solitary brain metastasis and high-grade glioma by diffusion tensor imaging. Br J Radiol 2005; 78:533 537 6. Law M, Cha S, Knopp EA, Johnson G, Arnett J, Litt AW. High-grade gliomas and solitary metastases: differentiation by using perfusion and proton spectroscopic MR imaging. Radiology 2002; 222:715 721 7. Yamasaki F, Kurisu K, Satoh K, et al. Apparent diffusion coefficient of human brain tumors at MR imaging. Radiology 2005; 235:985 991 8. Provenzale JM, Engelter ST, Petrella JR, Smith JS, MacFall JR. Use of MR exponential diffusionweighted images to eradicate T2 shine-through effect. AJR 1999; 172:537 539 9. Kono K, Inoue Y, Nakayama K, et al. The role of diffusion-weighted imaging in patients with brain tumors. AJNR 2001; 22:1081 1088 10. Krabbe K, Gideon P, Wagn P, Hansen U, Thomsen C, Madsen F. MR diffusion imaging of human intracranial tumors. Neuroradiology 1997; 39:483 489 11. Sugahara T, Korogi Y, Kochi M, 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 12. Kitis 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 13. Bulakbasi N, Kocaoglu M, Ors F, Tayfun C, Uçöz T. Combination of single-voxel proton MR spectroscopy and apparent diffusion coefficient calculation in the evaluation of common brain tumors. AJNR 2003; 24:225 233 14. Rollin N, Guyotat J, Streichenberger N, Honnorat J, Tran Minh VA, Cotton F. Clinical relevance of diffusion and perfusion magnetic resonance imaging in assessing intra-axial brain tumors. Neuroradiology 2006; 48:150 159 15. Stadnik TW, Chaskis C, Michotte A, et al. Diffusion-weighted MR imaging of intracerebral masses: comparison with conventional MR imaging and histologic findings. AJNR 2001; 22:969 976 16. Cha S. Update on brain tumor imaging: from anatomy to physiology. AJNR 2006; 27:475 487 17. Murakami R, Hirai T, Sugahara T, et al. Grading astrocytic tumors by using apparent diffusion coefficient parameters: superiority of a one- versus two-parameter pilot method. Radiology 2009; 251:838 845 18. Lee EJ, Lee SK, Agid R, Bae JM, Keller A, Terbrugge K. Preoperative grading of presumptive low-grade astrocytomas on MR imaging: diagnostic value of minimum apparent diffusion coefficient. AJNR 2008; 29:1872 1877 19. Higano S, Yun X, Kumabe T, et al. Malignant astrocytic tumors: clinical importance of apparent diffusion coefficient in prediction of grade and prognosis. Radiology 2006; 241:839 846 20. Murakami R, Sugahara T, Nakamura H, et al. Malignant supratentorial astrocytoma treated with postoperative radiation therapy: prognostic value of pretreatment quantitative diffusion-weighted MR imaging. Radiology 2007; 243:493 499 21. Eis M, Els T, Hoehn-Berlage M, Hossmann KA. Quantitative diffusion MR imaging of cerebral tumor and edema. Acta Neurochir Suppl (Wien) 1994; 60:344 346 22. Burtscher 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. AJNR 2000; 21:84 93 23. Sijens PE, Knopp MV, Brunetti A, et al. 1 H MR spectroscopy in patients with metastatic brain tumors: a multicenter study. Magn Reson Med 1995; 33:818 826 24. Poptani H, Gupta RK, Roy R, Pandey R, Jain VK, Chhabra DK. Characterization of intracranial mass lesions with in vivo proton MR spectroscopy. AJNR 1995; 16:1593 1603 25. Ishimaru H, Morikawa M, Iwanaga S, Kaminogo M, Ochi M, Hayashi K. Differentiation between high-grade glioma and metastatic brain tumor using single-voxel proton MR spectroscopy. Eur Radiol 2001; 11:1784 1791 26. Lu S, Ahn D, Johnson G, Law M, Zagzag D, Grossman RI. Diffusion-tensor MR imaging of intracranial neoplasia and associated peritumoral edema: introduction of the tumor infiltration index. Radiology 2004; 232:221 228 76 AJR:196, January 2011