Diffusion-weighted imaging predicts postoperative persistence in meningioma patients with peritumoural abnormalities on magnetic resonance imaging

Journal of Clinical Neuroscience (2003) 10(5), 589–593 0967-5868/$ - see front matter ª 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0967-5868(03)00093-6

Radiology study

Diffusion-weighted imaging predicts postoperative persistence in meningioma patients with peritumoural abnormalities on magnetic resonance imaging Akira Nakamizo MD, Takanori Inamura MD PHD, Shinya Yamaguchi MD, Satoshi Inoha MD, Toshiyuki Amano Kiyonobu Ikezaki MD PHD, Shunji Nishio MD PHD, Yasuhiko Nakamura BT, Masashi Fukui MD PHD

MD,

Department of Neurosurgery, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan

Summary Objective: While diffusion-weighted magnetic resonance imaging (MRI) has been used to study malignant brain tumours, this modality has not been used to study MRI abnormalities surrounding meningiomas. Methods: We examined intensity and apparent diffusion coefficient (ADC) on diffusion weighted imaging (DWI) for predicting postoperative persistence of MRI abnormalities surrounding meningiomas as well as characterizing the tumours. Results: Of 36 meningiomas who underwent gross total resection, 27 (75%) showed hyperintensity on DWI at b ¼ 1100 s=mm2 . No atypical meningiomas were hypointense on DWI. Of the 26 supratentorial meningiomas, 18 (69.0%) had associated MRI abnormality. No significant correlation was seen between tumour intensity on DWI and existence of surrounding MRI abnormality. Meningothelial meningiomas showed a relatively low prevalence of MRI abnormalities surrounding tumour (30%). Of 11 patients who underwent sequential MRI, all MRI abnormalities surrounding tumour showing isointensity and high ADC on preoperative DWI disappeared after surgery (from 3 weeks to 10 months). All MRI abnormalities surrounding tumour showing hyperintensity and low ADC on preoperative DWI persisted on final follow-up MRI (from 6 months to 20 months). Conclusion: The postoperative course of MRI abnormality surrounding tumour might be predictable from the intensity and ADC on preoperative DWI. Since MRI abnormalities associated with meningiomas can cause preoperative neurologic deficits. We hypothesise that abnormalities with restricted diffusion will be more likely to be associated with a preoperative deficit, and more likely to remain after removal of the causative meningioma. ª 2003 Elsevier Science Ltd. All rights reserved. Keywords: ADC, apparent diffusion coefficient, diffusion-weighted imaging, DWI, magnetic resonance imaging, meningioma, peritumoural brain oedema

INTRODUCTION

PATIENTS AND METHODS

Echo-planar diffusion-weighted imaging (DWI) is a relatively new technique that can generate images based on the molecular diffusion of water.12;13 DWI findings with meningiomas have been reported, although the detailed features of peritumoural brain oedema has not.21 Meningiomas are frequently accompanied by peritumoural brain oedema (PTBE) with or without gliosis despite the benign, extraaxial character of the tumour.1;11;18 PTBE and gliosis can in itself cause neurologic deficits in patients with meningioma. Like brain oedema in general, PTBE consists of two types, vasogenic and cytotoxic; vasogenic oedema is reversible, while cytotoxic oedema is thought to be irreversible.2;14 We hypothesised that DWI might be used to predict persistence of a postoperative neurologic deficit by distinguishing vasogenic oedema from cytotoxic oedema and gliosis. In this study, we examined and correlated preoperative findings on DWI in patients with meningioma with respect to tumour histology as well as MRI abnormality surrounding the tumour.

Patients

Received 14 January 2002 Accepted 28 January 2003 Correspondence to: Akira Nakamizo MD, Department of Neurosurgery, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan. Tel.: +81-92-642-5524; Fax: +81-92-642-5526; E-mail: [email protected]

Thirty-six patients with meningioma who underwent gross total resection were examined in this study, including 12 male and 24 female subjects ranging from 32 to 85 years old (mean
SD; 56:0
13:6). MRI Preoperative magnetic resonance imaging (MRI) was performed using a 1.5-Tesla magnetic field (Siemens Magnetom Vision, Germany). MRI sequences were as follows: T1-weighted imaging (T1WI: TR, 528 ms; TE, 17 ms; FOV, 23  23 cm; matrix, 256  256; slice thickness, 5 mm with a 2.5-mm gap); T2weighted imaging (T2WI: TR, 3000 ms; TE, 90 ms; FOV, 23  23 cm; matrix, 256  256; slice thickness, 5 mm with a 2.5mm gap); and DWI (TR, 0.8 ms; TE, 123 ms; FOV, 23  23 cm; matrix, 128  128; slice thickness, 5 mm with a 2.5-mm gap). Region of interest (ROI) analysis was performed on the axial images as described.5;22 The current hardware allowed diffusion gradients of up to 1.1 G/cm, with a rise time of 320 ls. The length of the diffusion gradients was 26 ms, and the diffusion-time was 60.1 ms. ROI was made from the center of the DWI abnormality. ADC maps were generated from 2 or more b value images by filtering the data on a pixel-by-pixel basis. The exponential function that describes the change in magnetic resonance signal intensity with the diffusion weighting b value is S ¼ S0 exp ðb  ADCÞ, where S and S0 are the pixel signal intensity with (S) and without (S0 ) diffusion weighting applied. A set of diffusion weighted images was evaluated in each of the 3 orthogonal directions (x; y; z). When diffusion information from all 589

Journal of Clinical Neuroscience (2003) 10(5), 589–593

18 (100%) 18 (100%) 13 (100%) DWI, diffusion-weighted imaging; MRI, magnetic resonance imaging; H, hyperintense; I, isointense; L, (hypointense); +, present; ), absent.

23 (100%) 8 (100%) 1 (100%) 1 (100%) 10 (100%) 2 (100%)

10 (56%) 3 (17%) 5 (27%) 17 (94%) 1 (6%) 0 9 (69%) 3 (23%) 1 (8%) 18 (78%) 1 (5%) 4 (17%) 7 (88%) 1 (12%) 0 0 1 (100%) 0 1 (100%) 0 0 6 (60%) 1 (10%) 3 (30%)

Secretory Microcystic

Histology

Transitional Fibrous

1 (50%) 0 1 (50%) 12 (86%) 1 (7%) 1 (7%)

14 (100%) 36 (100%) Total

Relationship between tumour intensity on DWI and presence of surrounding MRI abnormality In 26 supratentorial meningiomas, no significant correlation was noted between tumour intensity on DWI and presence of surrounding MRI abnormality (Table 3).

Meningothelial

Relationship between histologic type of tumour and presence of surrounding MRI abnormality Of 26 supratentorial meningiomas, 18 (69.0%) were accompanied by surrounding MRI abnormality. Only meningothelial meningiomas showed a lower occurrence rate for MRI abnormality (30%) than other types (Table 2). Among transitional meningiomas, 5 of 6 (83%) showed surrounding MRI abnormality; among fibrous meningiomas, 2 of 2; and among atypical meningiomas, 7 of 7.

36 Tumors

Relationship between histologic type and intensity of tumour on MRI No meningothelial meningiomas were hyperintense on T1WI, and no atypical or transitional meningiomas were hypointense on T2WI. Of 14 meningothelial meningiomas, 12 (86%) were hyperintense on DWI at b ¼ 1100 s=mm2 , as were 5 of 8 atypical meningiomas (88%; Table 1). No atypical meningiomas were hypointense on DWI at b ¼ 1100 s=mm2 . Other histologic types either showed various intensities or were too few in number for evaluation.

Tumor intensity in DWI

Histologic type Of 36 meningiomas, 14 were meningothelial, 2 were fibrous, 10 were transitional, 1 was microcystic, 1 was secretary, and 8 were atypical.

Table 1

Intensity of tumour on MRI Of 36 meningiomas, 18 meningiomas (50%) were hypointense and 12 (33%) were isointense on T1WI, while 28 meningiomas (78%) were hyperintense on T2WI. On DWI at b ¼ 1100 s=mm2 , 27 (75%) were hyperintense, while 5 (14%) were hypointense (Table 1).

Intensity of tumour on DWI and histology, presence of peritumoural MRI abnormality, and tumour consistency

Tumours

No. of cases (%)

RESULTS

27 (75%) 4 (11%) 5 (14%)

Atypical

Postoperatively, 11 patients were followed by sequential MRI examinations. In these subjects we examined the relationship between intensity or ADC of MRI abnormalities surrounding tumour on DWI and their persistence. Cases were grouped according to ADC on preoperative DWI, including a high ADC group (n ¼ 3, ADC > 1:50  103 mm2 =s) and a low ADC group (n ¼ 7, ADC < 1:50  103 mm2 =s; Table 4).

H I L

Postoperative findings

Soft

Tumor consistency Peritumoural MRI abnormality

We examined the relationship of tumour intensity on DWI to histologic tumour type, and to the presence of surrounding MRI abnormality. Experienced pathologists histologically examined all specimens. Since the cerebellum consists mainly of gray matter and infratentorial tumours were associated with relatively little MRI abnormality, we confined analyses concerning MRI abnormalities surrounding tumour to the 26 supratentorial meningiomas. Three cases treated with preoperative embolization were excluded.

)

Examination procedure

+

3 directions was available, ADC maps were generated for each direction and then averaged together to remove contrast due to diffusion anisotropy.

Hard

590 Nakamizo et al.

ª 2003 Elsevier Science Ltd. All rights reserved.

MRI abnormality surrounding meningioma 591

Table 2 Histologic type of tumour and presence of peritumoural MRI abnormality Histotype

for 20 months after surgery in the case with longest follow-up. Rise of ADC to more than 1:50  103 mm2 =s was associated with postoperative disappearance of MRI abnormality surrounding tumour (sensitivity 100%, specificity 100%; Fig. 3).

No. of cases (%) 26 Tumors

Peritumoural MRI abnormality )

+ Meningothelial Fibrous Transitional Microcystic Atypical

10 (38%) 2 (8%) 6 (23%) 1 (4%) 7 (27%)

Total

26 (100%)

3 2 5 1 7

(30%) (100%) (83%) (100%) (100%)

18 (69%)

DISCUSSION DWI

7 (70%) 0 1 (17%) 0 0

DWI is a relatively new MRI technique that detects minute motions of hydrogen nuclei in water molecules.13;12 Clinical applications have included detection and characterization of cerebral ischaemia, mapping of white matter development and anatomy, and distinguishing cystic and necrotic areas from solid portions of malignant brain tumours.6;15;16;20;21;23 Brain abscesses have been found to be strongly hyperintense on DWI; cystic or necrotic areas in a brain tumour, hypointense; and solid portions of high-grade gliomas and metastatic brain tumours, hyperintense.16;21

8 (31%)

MRI; magnetic resonance imaging; +, present; ), absent.

MRI abnormalities surrounding tumour Postoperative changes in MRI abnormality surrounding tumour Of 11 patients followed by sequential MRI examinations, their tumours included 1 meningothelial meningioma, 1 fibrous meningioma, 4 transitional meningiomas, and 5 atypical meningiomas (Table 4). The follow-up MRI was performed at intervals ranging from 3 weeks to 20 months. The mean ADC (
SD) for patients with high ADC on preoperative DWI was 2:14
0:71, while with low ADC was 1:22
0:19. Significant difference of ADC between a high ADC and a low ADC group was noted (Mann–Whitney U test; p ¼ 0:030; p < 0:05). In all cases of high ADC group, the disappearance of MRI abnormalities surrounding tumours ranged from 3 weeks to 10 months after surgery (Fig. 1). In contrast, all cases of low ADC showed persistence of MRI abnormalities surrounding tumour on final follow-up MRI (Fig. 2). Follow-up periods in this group ranged from 6 months to 20 months. MRI abnormality surrounding the tumour has persisted Table 3

Histologic type of tumour In the present study, 75% of meningiomas were hyperintense on DWI at b ¼ 1100 s=mm2 . Notably, no atypical meningioma showed hypointensity on DWI. Most atypical meningiomas were hyperintense, as are most malignant gliomas. These findings reflect the histologic findings of densely packed, often randomly organised tumour cells, which would be expected to inhibit motion of water molecules and restrict diffusion.12;21 Filippi et al.7 reported that atypical and malignant meningiomas tended to show markedly hyperintense on DWI, and benign meningiomas had a varied appearance on DWI except for densely calcified or psammomatous meningiomas. They supposed that the reduced extracellular water content of malignant and atypical meningiomas, and the histopathologic features, creating a complex, local environment that restricted the normal diffusion water molecules, caused

Intensity of supratentorial tumours on DWI and presence of peritumoural MRI abnormality

Tumor intensity in DWI

No. of cases (%) 26 Tumors

Histology

MRI abnormality

Meningothelial

Fibrous

Transitional

Microcystic

Atypical

+

)

H I L

20 (77%) 1 (4%) 5 (19%)

9 (90%) 0 1 (10%)

1 (50%) 0 1 (50%)

3 (50%) 0 3 (50%)

1 (100%) 0 0

6 (80%) 1 (20%) 0

13 (72%) 1 (6%) 4 (22%)

7 (88%) 0 1 (12%)

Total

26 (100%)

10 (100%)

2 (100%)

6 (100%)

1 (100%)

7 (100%)

18 (100%)

8 (100%)

DWI, diffusion-weighted imaging; MRI, magnetic resonance imaging; H, hyperintense; I, isointense; L, (hypointense); +, present; ), absent.

Table 4

Intentsity and ADC of peritumoural MRI abnormality on preoperative DWI and its persistence on postoperative MRI

No. of case

1 2 3 4 5 6 7 8 9 10 11

Histologic type

Preoperative DWI at b = 1100 s/mm2

ADC on preoperative DWI (·10)3 mm2/s)

Persistence of MRI abnormality

Meningothelial Fibrous Transitional Transitional Atypical Atypical Atypical Transitional Atypical Atypical Transitional

I I I I H H H H H H H

1.62 1.86 2.95 n.p. 1.47 1.39 1.26 1.18 0.90 1.12 1.25

3 weeks ()) 1.5 months ()) 6 months ()) 10 months ()) 12 months (+) 12 months (+) 19 months (+) 14 months (+) 6 months (+) 17 months (+) 20 months (+)

ADC, apparent diffusion coefficient; MRI, magnetic resonance imaging; DWI, diffusion-weighted imaging; +, present; ), absent.

ª 2003 Elsevier Science Ltd. All rights reserved.

Journal of Clinical Neuroscience (2003) 10(5), 589–593

592 Nakamizo et al.

Fig. 1 Case 2, pictured here, is a typical example where preoperative diffusion-weighted imaging (DWI) at b ¼ 1100 s=mm2 showed magnetic-resonanceimaging (MRI) abnormality surrounding tumour to be isointense. The patient was a 32-year-old man with a meningioma in the trigone of the left lateral ventricle. Upper, preoperative T1-weighted imaging (T1WI), T2-weighted imaging (T2WI), enhanced T1WI with administration of gadolinium contrast material (Gd), fluid attenuated inversion recovery imaging (FLAIR), DWI at b ¼ 300 s=mm2 (DWI b ¼ 300), and DWI at b ¼ 1100 s=mm2 (DWI b ¼ 1100); and lower, postoperative T1WI, T2WI, Gd, and FLAIR at 1.5 months after surgery. MRI abnormality surrounding tumour resolved after surgery.

Fig. 2 Case 10, pictured here, is a typical example where preoperative diffusion-weighted imaging (DWI) at b ¼ 1100 s=mm2 showed magnetic-resonanceimaging (MRI) abnormality surrounding tumour to be hyperintense. The patient was a 56-year-old man with a meningioma of the left sphenoid ridge. Upper, preoperative T1-weighted imaging (T1WI), T2-weighted imaging (T2WI), enhanced T1WI with administration of gadolinium contrast material (Gd), fluid attenuated inversion recovery imaging (FLAIR), DWI at b ¼ 1100 s=mm2 (DWI b ¼ 1100), and apparent diffusion coefficient (ADC) map; and lower, postoperative T1WI, T2WI, Gd, and FLAIR at 17 months after surgery. MRI abnormality surrounding tumour is seen to persist.

companying PTBE as MRI abnormality, with the transitional type having a significantly stronger association with PTBE.10 In other types, however, the frequency of accompanying PTBE is uncertain.4;9;10;17 Meningothelial meningiomas were accompanied by MRI abnormalities less frequently in our study than in other reports, possibly reflecting the small numbers of cases in our study. For other types of meningiomas, MRI abnormalities surrounding tumours frequency tended to agree with previous reports. We previously indicated that PTBE was influenced by blood supply from surrounding cerebral arteries,10 but we did not address that issue in the present study. We believe that several factors may affect formation of PTBE. Preoperative DWI and postoperative PTBE

Fig. 3 Scatterplot showing the difference of apparent diffusion coefficient (ADC; 103 mm2 =s) in two groups: Disappearance, a group of patients whose MRI abnormalities surrounding tumour disappeared postoperatively; Persistence, a group of patients whose MRI abnormalities surrounding tumour persisted postoperatively. The reference line shows an ADC of 1:50  103 mm2 =s.

this phenomenon. DWI may be used to differentiate various types of tumour tissue. Several authors have reported a relationship between histologic type of meningioma and the frequency of acJournal of Clinical Neuroscience (2003) 10(5), 589–593

MRI abnormalities are accompanied by meningiomas including PTBE and gliosis, however, they cannot be differentiated by conventional MRI. Brain oedema has been classified as either vasogenic or cytotoxic.3 In vasogenic oedema, which is caused by increased permeability of brain capillary endothelial cells, the interstitial fluid increases in volume. In contrast, cytotoxic oedema is characterised by increased intracellular water content with swelling of glia, neurons, and endothelial cells.8 Vasogenic oedema is reversible, while cytotoxic oedema has been suggested to be at least partially irreversible as with gliosis.8 On the other hand, DWI at high b values of 1000 s=mm2 can detect changes in the diffusion of water molecules that are specific to cytotoxic oedema, and can distinguish vasogenic oedema from ª 2003 Elsevier Science Ltd. All rights reserved.

MRI abnormality surrounding meningioma 593

cytotoxic oedema in the human brain.2;19 Previous studies have demonstrated that a reduction in the ADC is correlated with signal hyperintensity on DWI 2;19 . In the present study, MRI abnormalities surrounding tumour showing isointensity on DWI at b ¼ 1100 s=mm2 and ADC higher than 1:50  103 mm2 =s rapidly disappeared after tumour removal, while a preoperatively imaged MRI abnormalities surrounding tumour showing hyperintense on DWI and ADC lower than 1:50  103 mm2 =s persisted following surgery. These observations suggest that DWI can distinguish vasogenic brain oedema from cytotoxic brain oedema and gliosis related to meningiomas. As meningioma is a benign extraaxial tumour, no surgical specimen from the underlying brain was available for histopathologic study. Nonetheless, we demonstrated that intensity or ADC of MRI abnormalities surrounding tumour on preoperative DWI might be used to predict the postoperative course of MRI abnormalities. Since MRI abnormality associated with meningioma sometimes resulted in neurologic deficits, such predictions of the time course of MRI abnormality might be used to predict postoperative clinical improvement. For example, if the MRI abnormality surrounding tumour involving the motor region in a patient with hemiparesis was isointense and showed high ADC on preoperative DWI, hemiparesis might resolve relatively soon after surgery. On the other hand, if MRI abnormality surrounding tumour was hyperintense and showed low ADC on preoperative DWI, hemiparesis might be prolonged. Only 11 cases were examined in our study and the intervals for follow-up MRI not standardized. A large, precisely designed prospective study is needed; however, our preliminary results concerning DWI in the assessment of patients with meningioma, and associated MRI abnormality are promising and reflect the potential of this modality.

ACKNOWLEDGEMENTS This investigation was supported by Grant (12671365) from the Ministry of Education, Japan. We thank Ms Yoshie Hirosawa for her valuable assistance in preparing the manuscript. A part of this study was carried out at Morphology Core, Graduate School of Medical Sciences, Kyushu University. REFERENCES 1. Atkinson JL, Lane JI. Frontal sagittal meningioma: tumor parasitization of cortical vasculature as the etiology of peritumoral edema. J Neurosurg 1994; 81: 924–926. 2. Baird AE, Benfield A, Schlaug G, Siewert B, Lovblad KO, Edelman RR, Warach S. Enlargement of human cerebral ischemic lesion volumes measured by diffusion-weighted magnetic resonance imaging. Ann Neurol 1997; 41: 581–589. 3. Bell BA. A history of the study of cerebral edema. Neurosurgery 1983; 13: 724–728. 4. Bradac GB, Ferszt R, Bender A, Schorner W. Peritumoral edema in meningiomas. Neuroradiology 1986; 28: 304–312.

ª 2003 Elsevier Science Ltd. All rights reserved.

5. Brunberg JA, Chenevert TL, McKeever PE, Ross DA, Junck LR, Muraszko KM, Dauser R, Pipe JG, Betley AT. In vivo MR determination of water diffusion coefficients and diffusion anisotropy: correlation with structural alteration in gliomas of the cerebral hemispheres. Am J Neuroradiol 1995; 16: 361–371. 6. Chenevert TL, Brunberg JA, Pipe JG. Anisotropic diffusion in human white matter: demonstration with MR techniques in vivo. Radiology 1990; 177: 401–405. 7. Filippi CG, Edgar MA, Ulug AM, Prowda JC, Heier LA, Zimmerman RD. Appearance of meningiomas on diffusion-weighted images: correlating diffusion constants with histopathologic findings. Am J Neuroradiol 2001; 22: 65–72. 8. Fishman R. Brain edema and disorders of intracranial pressure. In: Rowland L (ed) Merrit’s Textbook of Neurology. Lea & Febiger, Philadepha London 1989; 262–271. 9. Go KG, Wilmink JT, Molenaar WM. Peritumoral brain edema associated with meningiomas. Neurosurgery 1988; 23: 175–179. 10. Inamura T, Nishio S, Takeshita I, Fujiwara S, Fukui M. Peritumoral brain edema in meningiomas – influence of vascular supply on its development. Neurosurgery 1992; 31: 179–185. 11. Kalkanis SN, Carroll RS, Zhang J, Zamani AA, Black PM. Correlation of vascular endothelial growth factor messenger RNA expression with peritumoral vasogenic cerebral edema in meningiomas. J Neurosurg 1996; 85: 1095–1101. 12. Le Bihan D, Breton E, Lallemand D, Grenier P, Cabanis E, Laval-Jeantet M. MR imaging of intravoxel incoherent motions: application to diffusion and perfusion in neurologic disorders. Radiology 1986; 161: 401–407. 13. Le Bihan D, Breton E, Lallemand D, Aubin ML, Vignaud J, Laval-Jeantet M. Separation of diffusion and perfusion in intravoxel incoherent motion MR imaging. Radiology 1988; 168: 497–505. 14. Loubinoux I, Volk A, Borredon J, Guirimand S, Tiffon B, Seylaz J, Meric P. Spreading of vasogenic edema and cytotoxic edema assessed by quantitative diffusion and T2 magnetic resonance imaging. Stroke 1997; 28: 419–426. 15. Marks MP, de Crespigny A, Lentz D, Enzmann DR, Albers GW, Moseley ME. Acute and chronic stroke: navigated spin-echo diffusion-weighted MR imaging. Radiology 1996; 199: 403–408. 16. Noguchi K, Watanabe N, Nagayoshi T, Kanazawa T, Toyoshima S, Shimizu M, Seto H. Role of diffusion-weighted echo-planar MRI in distinguishing between brain brain abscess and tumour: a preliminary report. Neuroradiology 1999; 41: 171–174. 17. Philippon J, Foncin JF, Grob R, Srour A, Poisson M, Pertuiset BF. Cerebral edema associated with meningiomas: possible role of a secretory-excretory phenomenon. Neurosurgery 1984; 14: 295–301. 18. Salpietro FM, Alafaci C, Lucerna S, Iacopino DG, Todaro C, Tomasello F. Peritumoral edema in meningiomas: microsurgical observations of different brain tumor interfaces related to computed tomography. Neurosurgery 1994; 35: 638–641, discussion 641-632. 19. Schaefer PW, Buonanno FS, Gonzalez RG, Schwamm LH. Diffusion-weighted imaging discriminates between cytotoxic and vasogenic edema in a patient with eclampsia. Stroke 1997; 28: 1082–1085. 20. Sorensen AG, Buonanno FS, Gonzalez RG, Schwamm LH, Lev MH, Huang-Hellinger FR, Reese TG, Weisskoff RM, Davis TL, Suwanwela N, Can U, Moreira JA, Copen WA, Look RB, Finklestein SP, Rosen BR, Koroshetz WJ. Hyperacute stroke: evaluation with combined multisection diffusion-weighted and hemodynamically weighted echo-planar MR imaging. Radiology 1996; 199: 391–401. 21. Tien RD, Felsberg GJ, Friedman H, Brown M, MacFall J. MR imaging of high-grade cerebral gliomas: value of diffusion-weighted echoplanar pulse sequences. Am J Roentgenol 1994; 162: 671–677. 22. Ulug AM, Beauchamp N, Bryan RN, van Zijl PC. Absolute quantitation of diffusion constants in human stroke. Stroke 1997; 28: 483–490. 23. Warach S, Gaa J, Siewert B, Wielopolski P, Edelman RR. Acute human stroke studied by whole brain echo planar diffusion-weighted magnetic resonance imaging. Ann Neurol 1995; 37: 231–241.

Journal of Clinical Neuroscience (2003) 10(5), 589–593