September 2002, VOLUME 179
NUMBER 3

Recommend & Share

September 2002, Volume 179, Number 3

Neuroradiology

Using Relative Cerebral Blood Flow and Volume to Evaluate the Histopathologic Grade of Cerebral Gliomas: Preliminary Results

+ Affiliations:
1 Department of Radiology, Asan Medical Center, University of Ulsan College of Medicine, 388-1 Poongnapdong, Songpa-gu, Seoul, 138-736, South Korea.

2 Department of Neurosurgery, Asan Medical Center, University of Ulsan College of Medicine, Seoul, 138-736, South Korea.

3 Department of Radiology, Armed Forces Capital Hospital, San 13-4 Yooldong, Boondang-gu, Seongnam, Seoul, 463-040, South Korea.

4 Department of Information and Statistics, Daejeon University, 96-3 Yongwoo-dong, Daejeon, 300-716, Korea.

Citation: American Journal of Roentgenology. 2002;179: 783-789. 10.2214/ajr.179.3.1790783

ABSTRACT
Next section

OBJECTIVE. Relative cerebral blood flow has rarely been studied as part of the preoperative assessment of tumor grade, although relative cerebral blood volume is known to be useful for this assessment. The purpose of our study was to determine the usefulness of relative cerebral blood flow in assessing the histopathologic grade of cerebral gliomas.

SUBJECTS AND METHODS. MR imaging was performed in 17 patients with proven cerebral gliomas (11 high-grade gliomas and six low-grade gliomas), using a first-pass gadopentetate dimeglumine—enhanced T2-weighted echoplanar perfusion sequence. The perfusion data were deconvoluted with an arterial input function, using singular value decomposition to obtain a color map of relative cerebral blood volume and flow; the relative cerebral blood volume and flow ratios were expressed relative to values measured in the contralateral white matter. The Wilcoxon's rank sum test was performed to test the difference between the mean of the relative cerebral blood volume (or flow) ratio in high-grade gliomas and that in low-grade gliomas. Receiver operating characteristic curve analysis was used to evaluate the association between the relative cerebral blood volume (or flow) ratio and the grade of the glioma, as well as to calculate the relative cerebral blood volume and flow ratio cutoff value permitting discrimination between high- and low-grade gliomas. The correlation between relative cerebral blood volume and flow ratios was evaluated using Spearman's rank correlation analysis. We also made a qualitative assessment regarding the match or mismatch of areas of maximal contrast enhancement with the areas of highest color perfusion maps.

RESULTS. The mean of the relative cerebral blood volume ratio was 4.91 in the high-grade gliomas and 2.00 in the low-grade gliomas. The mean relative cerebral blood flow ratio was 4.82 in the high-grade gliomas and 1.83 in the low-grade gliomas. A significant difference in each relative cerebral blood volume and flow ratio was found between the high- and low-grade gliomas (Wilcoxon's rank sum test, p < 0.05). Both the relative cerebral blood volume and flow ratios strongly matched the grade of the glioma, but the difference between the two areas was not significant (receiver operating characteristic curve analysis, p > 0.05). The desired cutoff value was 2.93 in the relative cerebral blood volume ratio and 3.57 in the relative cerebral blood flow ratio. Additionally, there was a strong correlation between the relative cerebral blood volume and flow ratios (Spearman's rank correlation coefficient = 0.762; p < 0.05). There was frequent mismatch (33%) between the qualitative assessment of the contrast-enhanced T1-weighted MR images and the perfusion maps.

CONCLUSION. First-pass gadopentetate dimeglumine—enhanced T2-weighted echoplanar perfusion MR imaging is useful for the preoperative assessment of tumor grade. A relative cerebral blood flow ratio, in addition to a relative cerebral blood volume ratio, can be a useful tool in the evaluation of the histopathologic grade of cerebral gliomas.

Introduction
Previous sectionNext section

Gliomas are the most common primary neoplasms of the brain in adults. In patients with gliomas, therapeutic approaches, response to therapy, and prognosis differ considerably according to tumor grade. Therefore, finding the highest grade portion of the tumor to be biopsied is important but can be challenging. In addition, because of the frequent heterogeneity of gliomas, an unfortunate choice of biopsy site or one that is too small may result in the incorrect assignment of a low tumor grade and consequent misdetermination of optimal treatment strategies [1]. The two most important factors in determining the malignancy of gliomas are their ability to infiltrate the brain parenchyma and to recruit, synthesize, and proliferate vascular networks for further growth. Therefore, the direct measurement of angiogenesis by activated endothelial cells is a primary grading criterion for determining the biologic aggressiveness and histopathologic grading of gliomas [2,3,4].

Important limitations exist with conventional MR imaging techniques in the differentiation of low- and high-grade gliomas because conventional MR images may not show the most malignant tumor areas [1, 5], although, in general, a greater likelihood exists for contrast enhancement of high-grade tumors [6, 7].

Perfusion MR imaging consists of several recently developed MR techniques to noninvasively measure cerebral perfusion via the assessment of various hemodynamic measurements, such as cerebral blood volume, cerebral blood flow, and mean transit time [8]. Numerous studies emphasize the usefulness of the relative cerebral blood volume for intraaxial cerebral tumors or strokes [1, 2, 9,10,11]. However, relative cerebral blood flow has rarely been studied as part of the preoperative assessment of tumor grade [11], although several studies report the usefulness of the relative cerebral blood flow for managing strokes [9, 12]. Changes in cerebral blood volume and cerebral blood flow are known to be typically well correlated in that when cerebral blood volume rises, cerebral blood flow also rises, and vice versa [13]. Therefore, we hypothesized that relative cerebral blood flow might have a role comparable to that of relative cerebral blood volume in the characterization of the grade of cerebral gliomas. The purpose of our study was to determine the usefulness of relative cerebral blood flow in assessing the histopathologic grade of cerebral gliomas.

Subjects and Methods
Previous sectionNext section
Study Group

From December 1999 to July 2000, we prospectively investigated 17 consecutive patients (eight men and nine women) who were subsequently identified as having histopathologically proven gliomas. Only initial preoperative perfusion MR studies in patients with pathologically confirmed gliomas were included in our study. Therefore, postoperative or postradiation therapy perfusion MR studies were excluded. There were 11 high-grade gliomas and six low-grade gliomas. Patient ages ranged from 17 to 67 years (average age, 40.1 years). Study approval was obtained from the Asan Medical Center review board, and informed consent was obtained from all patients. Before imaging, we inserted an 18- or 20-gauge IV catheter in the antecubital area for contrast agent administration.

Imaging System and Sequences

Cerebral perfusion studies were performed using dynamic tracking of a bolus of gadopentetate dimeglumine (Magnevist; Schering, Berlin, Germany) administered IV. Images were acquired using a first-pass gadopentetate dimeglumine—enhanced T2-weighted echoplanar image sequence (TR/TE, 1200/42.1; flip angle, 90°; field of view, 250 × 250 mm; slice thickness, 6 mm; matrix, 128 × 128; pixel size, 1.95 × 1.95 mm) followed by conventional contrast-enhanced axial and coronal T1-weighted images (490/14) on a 1.5-T MR scanner (Magnetom Vision; Siemens Medical Systems, Erlangen, Germany). The location and size of the tumor and the subsequent positioning of the superior and inferior margins of the scan area of perfusion MR imaging were determined from the initially acquired T2-weighted MR images (3500/99). Images from five slices with 40 time frames per slice were obtained at 2-sec intervals. The first five acquisitions were performed before contrast injection to establish a precontrast baseline. On the fifth acquisition, gadopentetate dimeglumine (0.2 mmol/kg) was injected manually at a rate of about 4 mL/sec through an 18- or 20-gauge IV catheter and was followed immediately by a 10-mL saline flush.

Data Processing and Calculation of the Relative Cerebral Blood Volume and Flow Ratios

All processing was performed on an x86-based PC using software developed in Interactive Data Language programming (Research Systems, Boulder, CO).

Dynamic susceptibility contrast-enhanced MR imaging makes it possible to assess regional cerebral hemodynamics by analyzing signal intensity changes after the first pass of a paramagnetic contrast agent [11, 14, 15]. “Cerebral blood volume” refers to the amount of blood in a given region of brain tissue at any time, commonly measured in milliliters per 100 g of brain tissue, whereas “cerebral blood flow” refers to the amount of blood passing through a given region of brain tissue per unit of time, commonly measured in milliliters per 100 g of brain tissue per minute. In a first step, the relative cerebral blood volume map was derived on a pixel-by-pixel basis from the dynamic image sets. Before the starting point of the first-pass circulation, a representative number of baseline points were selected, and their average was calculated for each pixel as a baseline measure for signal intensity (S0). On a pixel-by-pixel basis, we converted the signal intensity (S) to changes in the T2* relaxation rate (Δ R2*), assuming an exponential relationship between the relative signal reduction and the local contrast medium concentration using the following equation [16]:

The relative cerebral blood volume map was generated by the numeric integration of the relative concentration (Δ R2*) of the first-pass bolus through each voxel on the basis of the kinetic principles of nondiffusible radionuclide [10, 17]. To estimate the first-pass concentration, we fitted a gamma-variate function to the time—signal course of each pixel. One popular approach for determining the relative cerebral blood flow by measuring the concentration remaining in tissue uses the area-to-height relationship, which states that the ratio of relative cerebral blood volume to relative cerebral blood flow equals the ratio of the area under the tissue concentration—time curve to the maximal height of the curve (Cmax) [18]. Administration of a real bolus of contrast material is always of finite duration, and the measured tissue concentration—time curve, Cm(t), is thus a convolution of the tissue concentration—time response to an idealized bolus, C(t), and the arterial input function (AIF) (t) [18]:

To obtain the area-to-height ratio, one must perform deconvolution analysis. To obtain an AIF, we manually positioned the 20 × 20 pixel region-of-interest box on the area covering the M2 portion of the middle cerebral artery at which an AIF is to be obtained (Fig. 1). However, the most suitable AIF was estimated automatically from only a single pixel in the region of interest, which showed an early and large increase in R2 after contrast injection. This means the earlier and larger signal change, consistent with the higher concentration of contrast material in the artery than that in the capillary bed of the gray matter, in the concentration—time curve in the artery (i.e., AIF). Then, the measured tissue concentration—time curve was deconvoluted with the AIF using the singular value decomposition method.

figure
View larger version (202K)
Fig. 1. Axial T1-weighted MR image. Region of interest of 20 × 20 pixel size was manually positioned on area covering M2 portion of middle cerebral artery to obtain arterial input function (AIF). Scanning region-of-interest area (arrow), program calculates most suitable AIF and marks point (asterisk).

The values thus obtained were calculated for each pixel and were used to generate a color overlay for the base images, resulting in the relative cerebral blood volume and flow maps. We defined high color value as color value higher than that of cerebral cortex, low color value as color value lower than that of cerebral cortex, and moderate color value as nearly equal to that of cerebral cortex. On each relative cerebral blood volume and flow map, a rectangular region of interest was manually positioned in the highest color levels of the solid portion of a tumor for measurement of relative cerebral blood volume and flow. The size of the region of interest was 10 × 10 mm and included at least 16 pixels. However, the relative cerebral blood volume and flow values obtained using this method were not absolute quantities; therefore, we normalized the results by expressing ratios relative to values measured in the contralateral white matter—that is, the relative cerebral blood volume (or flow) of a tumor divided by that of white matter (Fig. 2A,2B,2C,2D). We obtained the relative cerebral blood volume and flow ratios of each lesion after repeating this procedure three times and having the three values averaged by two radiologists.

figure
View larger version (197K)
Fig. 2A. 63-year-old woman with glioblastoma multiforme in right frontal lobe. Contrast-enhanced T1-weighted MR image shows irregular strong enhancement of mass (arrows).

figure
View larger version (150K)
Fig. 2B. 63-year-old woman with glioblastoma multiforme in right frontal lobe. On relative cerebral blood volume (B) and cerebral blood flow (C) maps, there are correspondingly high cerebral blood volume and cerebral blood flow color values, respectively (arrows). Note match between perfusion maps (B and C) and contrast-enhanced T1-weighted MR images (A). Maximal relative cerebral blood volume and relative cerebral blood flow ratios are also high at 4.82 and 5.85, respectively.

figure
View larger version (156K)
Fig. 2C. 63-year-old woman with glioblastoma multiforme in right frontal lobe. On relative cerebral blood volume (B) and cerebral blood flow (C) maps, there are correspondingly high cerebral blood volume and cerebral blood flow color values, respectively (arrows). Note match between perfusion maps (B and C) and contrast-enhanced T1-weighted MR images (A). Maximal relative cerebral blood volume and relative cerebral blood flow ratios are also high at 4.82 and 5.85, respectively.

figure
View larger version (128K)
Fig. 2D. 63-year-old woman with glioblastoma multiforme in right frontal lobe. Relative cerebral blood flow map shows placement of rectangular regions of interest for measurement of relative cerebral blood flow in tumor (GM) and in contralateral peritrigonal white matter (WM). Relative cerebral blood flow ratio was relative cerebral blood flow of tumor divided by that of white matter.

Image Analysis and Statistics

The mean and standard deviations of the relative cerebral blood volume and flow ratios were calculated in each high- and low-grade glioma. The Wilcoxon's rank sum test (two-tailed) was performed to test the significance of the difference between the mean or relative cerebral blood volume (or flow) ratio in high-grade gliomas and that in low-grade gliomas. Receiver operating characteristic (ROC) curve analysis was used to evaluate the strength of the association between the relative cerebral blood volume (flow) ratio and the grade of the glioma (Medcalc Ver. 4.2, Mariakerke, Belgium). In this instance, the ROC curve was created as follows: the true-positive rate (sensitivity) and the false-positive rate (1 —specificity) were paired across all potential cutoff points, permitting discrimination between high- and low-grade gliomas. In the ROC curve for the relative cerebral blood volume (or flow) ratio, the size of the area under the ROC curve presented the intensity of the association between the relative cerebral blood volume (or flow) ratio and the grade of the glioma: the closer to 1, the stronger the association; the closer to 0.5, the weaker the association.

A desired cutoff value should have high true-positive and low false-positive rates for partitioning high-grade from low-grade gliomas. Each cutoff value for the relative cerebral blood volume and flow ratios was determined by assuming equal misclassification rates. Correlation between the relative cerebral blood volume and flow ratios was evaluated using Spearman's correlation coefficient. A difference of less than 0.05 in a p value was considered statistically significant.

Qualitative assessment was made regarding the match or mismatch of areas of maximal contrast enhancement on the contrast-enhanced T1-weighted MR images with areas of highest color on the relative cerebral blood volume and flow maps.

Results
Previous sectionNext section

Of the 17 patients studied, 11 had high-grade gliomas and six had low-grade gliomas. Of these 17 patients, four underwent stereotactic biopsy, and 13 underwent resection. All patients tolerated the perfusion echoplanar imaging sequence without any adverse reaction to the rapid bolus injection of contrast agent.

Magnetic susceptibility artifacts inherent in echoplanar imaging were prominent at bone—air interfaces, such as the petrous temporal bone, skull base, and paranasal sinuses. However, no lesion included in this study was markedly distorted by these artifacts.

The findings of the conventional MR imaging and perfusion maps from all patients were analyzed. Most of the high-grade gliomas (10/11 patients) showed moderate to strong enhancement, and the remaining lesion showed absent or minimal enhancement, whereas in the low-grade gliomas, the enhancement patterns were diverse—that is, strong (1), moderate (1), mild (1), and absent or minimal (3). Figures 2A,2B,2C,2D and 3A,3B,3C show examples of high-grade gliomas. The one case of a low-grade glioma showing strong enhancement on contrast-enhanced T1-weighted MR images was the pilocytic astrocytoma identified at pathology (Fig. 4A,4B,4C). The extent of peritumoral T2 abnormality was greater in high-grade gliomas. Necrosis on MR images was shown only in high-grade gliomas in seven of the 11 patients (Fig. 3A,3B,3C). In the high-grade gliomas, the relative cerebral blood volume maps had high color values in most of our patients (10/11), and the relative cerebral blood flow maps showed similar results, although two patients had moderate color values instead of high color values. In the low-grade gliomas, the color values of the relative cerebral blood volume maps were diverse with high (2), moderate (1), and low (3) values, and the relative cerebral blood flow maps showed similar results.

figure
View larger version (190K)
Fig. 3A. 55-year-old woman with glioblastoma multiforme in right temporal lobe. Contrast-enhanced T1-weighted MR image shows strong enhancement of entire rim of mass (arrows).

figure
View larger version (148K)
Fig. 3B. 55-year-old woman with glioblastoma multiforme in right temporal lobe. On relative cerebral blood volume (B) and cerebral blood flow (C) maps, there are focal high cerebral blood volume and cerebral blood flow color values in medial portion (arrows) of tumor and low cerebral blood volume and cerebral blood flow color values in other portion of tumor. Note mismatch between perfusion maps (B and C) and contrast-enhanced T1-weighted MR image (A). Maximal relative cerebral blood volume and relative cerebral blood flow ratios are high at 5.48 and 4.58, respectively, in medial portion of tumor.

figure
View larger version (146K)
Fig. 3C. 55-year-old woman with glioblastoma multiforme in right temporal lobe. On relative cerebral blood volume (B) and cerebral blood flow (C) maps, there are focal high cerebral blood volume and cerebral blood flow color values in medial portion (arrows) of tumor and low cerebral blood volume and cerebral blood flow color values in other portion of tumor. Note mismatch between perfusion maps (B and C) and contrast-enhanced T1-weighted MR image (A). Maximal relative cerebral blood volume and relative cerebral blood flow ratios are high at 5.48 and 4.58, respectively, in medial portion of tumor.

figure
View larger version (186K)
Fig. 4A. 50-year-old man with low-grade pilocytic astrocytoma in right basal ganglia. Contrast-enhanced T1-weighted MR image shows heterogeneous strong enhancement of mass (arrows).

figure
View larger version (151K)
Fig. 4B. 50-year-old man with low-grade pilocytic astrocytoma in right basal ganglia. On relative cerebral blood volume (B) and cerebral blood flow (C) maps, there are low cerebral blood volume and cerebral blood flow color values, respectively (arrows). Note mismatch between perfusion maps (B and C) and contrast-enhanced T1-weighted MR image (A). Maximal relative cerebral blood volume and relative cerebral blood flow ratios are also low, at 2.44 and 1.36, respectively.

figure
View larger version (147K)
Fig. 4C. 50-year-old man with low-grade pilocytic astrocytoma in right basal ganglia. On relative cerebral blood volume (B) and cerebral blood flow (C) maps, there are low cerebral blood volume and cerebral blood flow color values, respectively (arrows). Note mismatch between perfusion maps (B and C) and contrast-enhanced T1-weighted MR image (A). Maximal relative cerebral blood volume and relative cerebral blood flow ratios are also low, at 2.44 and 1.36, respectively.

The relative cerebral blood volume ratio in the high-grade gliomas (n = 11) varied from 0.87 to 7.88 with a mean of 4.91 ± 1.81 (SD), whereas in the low-grade gliomas (n = 6), it varied from 0.72 to 5.10 with a mean of 2.00 ± 1.54. The relative cerebral blood flow ratio in the high-grade gliomas (n = 11) varied from 1.33 to 11.22 with a mean of 4.82 ± 2.64; in the low-grade gliomas (n = 6), it varied from 0.82 to 3.42 with a mean of 1.83 ± 0.98. A significant difference existed in each relative cerebral blood volume and flow ratio between the high- and the low-grade gliomas based on the two-tailed Wilcoxon's rank sum test (p < 0.05). Both relative cerebral blood volume and flow ratios had a strong association with the grade of glioma. The areas under the ROC curve were 0.864 and 0.879 for the relative cerebral blood volume and flow ratios, respectively, and there was no significant difference in the area under the curve for either the relative cerebral blood volume or the relative cerebral blood flow ratio (p > 0.05) (Fig. 5). On the basis of equal misclassification rates, a cutoff value of 2.93 for the relative cerebral blood volume ratio (sensitivity, 90.9%; specificity, 83.3%) and a cutoff value of 3.57 for the relative cerebral blood flow ratio (sensitivity, 72.7%; specificity, 100%) best discriminated the high- and low-grade gliomas. Furthermore, there was a strong correlation between the relative cerebral blood volume and flow ratios with the high Spearman's rank correlation coefficient of 0.762 (p < 0.05) (Fig. 6).

figure
View larger version (26K)
Fig. 5. Receiver operating characteristic curves for relative cerebral blood volume (dashed line) and relative cerebral blood flow (solid line) ratios. Area under curve is 0.864 for relative cerebral blood volume ratio and 0.879 for relative cerebral blood flow ratio, with no significant difference between the two (p > 0.05).

figure
View larger version (21K)
Fig. 6. Scatterplot shows high correlation between relative cerebral blood volume and relative cerebral blood flow ratios. Line represents linear regression between relative cerebral blood volume and relative cerebral blood flow ratios. Spearman's correlation coefficient = 0.762, p < 0.05.

In the qualitative assessment of the contrast-enhanced T1-weighted MR images and the perfusion maps, a match occurred between the strongest enhanced area on contrast-enhanced T1-weighted MR images and the highest color value area on the perfusion maps in eight (67%) of the 12 patients showing moderate to strong enhancement on conventional contrast-enhanced T1-weighted MR images (Fig. 2A,2B,2C,2D). This result indicates that all the strongest enhanced areas on contrast-enhanced T1-weighted MR images corresponded to the highest color value areas on the perfusion maps. However, a mismatch occurred between the strongest enhanced area on contrast-enhanced T1-weighted MR images and the highest color value area on the perfusion maps in four (33%) of the 12 patients showing moderate to strong enhancement on conventional contrast-enhanced T1-weighted MR images (Figs. 3A,3B,3C and 4A,4B,4C). This result indicates that part of the most strongly enhanced area on contrast-enhanced T1-weighted MR images represented the highest color value area on the perfusion maps.

Discussion
Previous sectionNext section

To plan the optimal treatment strategy for cerebral gliomas, accurate determination of tumor grade is critical; in most histologic grading systems, vascular proliferation of gliomas is a diagnostic criterion for malignancy.

Recent developments in MR imaging have made it possible to assess perfusion abnormalities, with particular focus on relative cerebral blood volume in cerebral gliomas [1, 2, 19]. Knopp et al. [2] and Aronen et al. [1] each showed the statistically significant differences in relative cerebral blood volume between high- and low-grade gliomas, Knopp et al. using gradient-echo echoplanar imaging and Aronen et al. using spin-echo echoplanar imaging. The relative cerebral blood volume is clearly related to perfusion, and its values can be used to analyze the grades of gliomas. This rationale is based on the fact that “angiogenesis,” the proliferation of capillary endothelial cells, is one of the most important parameters for determining the histopathologic grade of gliomas along with the degree of cellular or nuclear pleomorphism and the presence or absence of necrosis [20, 21]. Furthermore, the median survival of patients with treated high-grade gliomas has been shown to be inversely related to the degree of endothelial proliferation of the original tumor [22]. Additionally, relative cerebral blood flow can play a role comparable to that of relative cerebral blood volume in the pathologic grading of cerebral gliomas.

Differentiation of high- and low-grade gliomas is sometimes difficult on conventional T1-weighted MR images alone. In our study, two of six low-grade gliomas showed moderate to strong contrast enhancement; this finding is usual in high-grade gliomas on conventional contrast-enhanced T1-weighted MR images. Most cases of high-grade gliomas showed moderate to strong contrast enhancement on conventional contrast-enhanced T1-weighted MR images. On the other hand, a statistically significant difference was found in each relative cerebral blood volume and flow ratio between high- and low-grade gliomas. There was, however, no statistically significant difference between the relative cerebral blood volume and flow ratios in either high- or low-grade gliomas. These statistical results suggest that a certain cutoff value of the relative cerebral blood volume and flow ratios can discriminate between high- and low-grade gliomas. This calculated cutoff value was 2.93 in the relative cerebral blood volume ratio and 3.57 in the relative cerebral blood flow ratio. We think that these two cutoff values could be useful in predicting glioma grade, considering its high sensitivity and specificity based on the results of the ROC curve. In particular, the specificity of the cutoff value of the relative cerebral blood flow ratio, 3.57, was 100%. Moreover, calculation of the relative cerebral blood volume and flow ratios is more objective and credible than visual assessment of the relative cerebral blood volume and flow maps.

Of the 12 tumors showing a moderate to strong degree of enhancement on contrast-enhanced T1-weighted MR images, four (33%) showed a mismatch between the most strongly enhanced area on contrast-enhanced T1-weighted MR images and the area of the highest color value on the relative cerebral blood volume and flow maps. In all four of these mismatched cases, the most strongly enhanced area on contrast-enhanced T1-weighted MR images was larger than the highest color value area on the perfusion maps. This observation that the area of the greatest vascular hyperplasia is smaller than the most strongly enhanced area on contrast-enhanced T1-weighted MR images, is not surprising because the contrast enhancement represents areas of contrast accumulation caused by pathologic alteration of the blood-brain barrier with or without concomitant vascular hyperplasia [2]. This breakdown of the blood—brain barrier can result from destruction of normal capillaries by the neoplastic process or from the pathologic structure of the vascular walls of newly formed abnormal capillaries, whereas the degree of perfusion abnormality reflects the degree of tumor microvascularity with or without destruction of the blood—brain barrier [2, 23].

A limitation of our study is the small number of patients. Another limitation is that total tumor resection was only performed in 12 of the 17 patients, and a one-to-one correlation was not performed between the resected pathologic specimen and the abnormal portion on conventional MR or perfusion images. Actually, high-grade gliomas could have been mistaken for low-grade gliomas because of sampling error [1]. Additional studies, including more study subjects and detailed histopathologic correlation of gliomas with the imaging findings, will be necessary to validate these results.

In conclusion, first-pass gadopentetate dimeglumine—enhanced T2-weighted echoplanar perfusion imaging is useful for the preoperative assessment of tumor grade. A relative cerebral blood flow ratio, in addition to a relative cerebral blood volume ratio, can be a useful tool for the discrimination of high- and low-grade cerebral gliomas.

Address correspondence to H. K. Lee.

We thank Bonnie Hami, Department of Radiology, University Hospitals of Cleveland, for her editorial assistance.

References
Previous sectionNext section
1. Aronen HJ, Gazit IE, Louis DN, et al. Cerebral blood volume maps of gliomas: comparison with tumor grade and histologic findings. Radiology 1994; 191:41-51 [Google Scholar]
2. Knopp EA, Cha S, Johnson G, et al. Glial neoplasms: dynamic contrast-enhanced T2*-weighted MR imaging. Radiology 1999; 211:791-798 [Google Scholar]
3. Zagzag D, Friedlander DR, Dosik J, et al. Tenascin-C expression by angiogenic vessels in human astrocytomas and by human brain endothelial cells in vitro. Cancer Res 1996; 56:182-189 [Google Scholar]
4. Brem S. The role of vascular proliferation in the growth of brain tumors. Clin Neurosurg 1976; 23:440-453 [Google Scholar]
5. Just M, Thelen M. Tissue characterization with T1, T2, and proton density values: results in 160 patients with brain tumors. Radiology 1988; 169:779-785 [Google Scholar]
6. Brant-Zawadzki M. Pitfalls of contrast-enhanced imaging in the nervous system. Magn Reson Med 1991; 22:243-248 [Google Scholar]
7. Brant-Zawadzki M, Berry I, Osaki L, Brasch R, Murovic J, Norman D. Gd-DTPA in clinical MR of the brain. 1. Intraaxial lesions. AJR 1986; 147:1223-1230 [Abstract] [Google Scholar]
8. Petrella JR, Provenzale JM. MR perfusion imaging of the brain: techniques and applications. AJR 2000; 175:207-219 [Abstract] [Google Scholar]
9. Sorensen AG, Copen WA, Ostergaard L, et al. Hyperacute stroke: simultaneous measurement of relative cerebral blood volume, relative cerebral blood flow, and mean tissue transit time. Radiology 1999; 210:519-527 [Google Scholar]
10. Sugahara T, Korogi Y, Kochi M, et al. Correlation of MR imaging—determined cerebral blood volume maps with histologic and angiographic determination of vascularity of gliomas. AJR 1998; 171:1479-1486 [Abstract] [Google Scholar]
11. Edelman RR, Mattle HP, Atkinson DJ, et al. Cerebral blood flow: assessment with dynamic contrast-enhanced T2*-weighted MR imaging at 1.5 T. Radiology 1990; 176:211-220 [Google Scholar]
12. Frilik AD, Rubin G, Yonas H, Wechsler LR. Relation between cerebral blood flow and neurologic deficit resolution in acute ischemic stroke. Neurology 1998; 51:177-182 [Google Scholar]
13. Grubb RL Jr, Raichle ME, Eichling JO, Ter-Pogossian MM. The effects of changes in PaCO2 on cerebral blood volume, blood flow, and vascular mean transit time. Stroke 1974; 5:630-639 [Google Scholar]
14. Albert MS, Huang W, Lee JH, Patlak CS, Springer CS Jr. Susceptibility changes following bolus injections. Magn Reson Med 1993; 29:700-708 [Google Scholar]
15. Rosen BR, Belliveau JW, Buchbinder BR, et al. Contrast agents and cerebral hemodynamics. Magn Reson Med 1991; 19:285-292 [Google Scholar]
16. Rosen BR, Belliveau JW, Vevea JM, Brady TJ. Perfusion imaging with NMR contrast agents. Magn Reson Med 1990; 14:249-265 [Google Scholar]
17. Axel L. Cerebral blood flow determination by rapid-sequence computed tomography: theoretical analysis. Radiology 1980; 137:679-686 [Google Scholar]
18. Rempp KA, Brix G, Wenz F, Becker CR, Guckel F, Lorenz WJ. Quantification of regional cerebral blood flow and volume with dynamic susceptibility contrast-enhanced MR imaging. Radiology 1994; 193:637-641 [Google Scholar]
19. Le Bihan D, Douek P, Argyropoulou M, Turner R, Patronas N, Fulham M. Diffusion and perfusion magnetic resonance imaging in brain tumors. Top Magn Reson Imaging 1993; 5:25-31 [Google Scholar]
20. Burger PC. Malignant astrocytic neoplasms: classification, pathologic anatomy, and response to treatment. Semin Oncol 1986; 13:16-26 [Google Scholar]
21. Brem S, Cotran R, Folkman J. Tumor angiogenesis: a quantitative method for histologic grading. J Natl Cancer Inst 1972; 48:347-356 [Google Scholar]
22. Fulling KH, Garcia DM. Anaplastic astrocytoma of the adult cerebrum: prognostic value of histologic features. Cancer 1985; 55:928-931 [Google Scholar]
23. Grossman RI, Wolf G, Biery D, et al. Gadolinium enhanced nuclear magnetic resonance images of experimental brain abscess. J Comput Assist Tomogr 1984; 8:204-207 [Google Scholar]

Recommended Articles

Using Relative Cerebral Blood Flow and Volume to Evaluate the Histopathologic Grade of Cerebral Gliomas: Preliminary Results

Full Access,
American Journal of Roentgenology. 2000;175:207-219. 10.2214/ajr.175.1.1750207
Citation | Full Text | PDF (13275 KB) | PDF Plus (13407 KB) 
Full Access, , , , , , , , , ,
American Journal of Roentgenology. 1998;171:1479-1486. 10.2214/ajr.171.6.9843274
Abstract | PDF (1535 KB) | PDF Plus (1174 KB) 
Full Access, , , , , , , , , ,
American Journal of Roentgenology. 2013;200:24-34. 10.2214/AJR.12.9543
Abstract | Full Text | PDF (1213 KB) | PDF Plus (1091 KB) 
Full Access, , , , , , , , , ,
American Journal of Roentgenology. 2013;201:W495-W510. 10.2214/AJR.12.9544
Abstract | Full Text | PDF (1776 KB) | PDF Plus (1515 KB) 
Full Access, , , ,
American Journal of Roentgenology. 2002;178:711-716. 10.2214/ajr.178.3.1780711
Abstract | Full Text | PDF (966 KB) | PDF Plus (352 KB) 
Full Access, , , ,
American Journal of Roentgenology. 2012;198:19-26. 10.2214/AJR.11.7417
Abstract | Full Text | PDF (1004 KB) | PDF Plus (980 KB)