J Neurosurg 107:873–877, 2007

Primary granulomatous angiitis of the central nervous
system: findings of magnetic resonance spectroscopy
and fractional anisotropy in diffusion tensor
imaging prior to surgery
Case report
TAKAAKI BEPPU, M.D.,1 TAKASHI INOUE, M.D.,1 HIDEAKI NISHIMOTO, M.D.,1
SHINICHI NAKAMURA, M.D., PH.D.,2 YOICHI NAKAZATO, M.D., PH.D.,3
KUNIAKI OGASAWARA, M.D.,1 AND AKIRA OGAWA, M.D., PH.D.1
Departments of 1Neurosurgery and 2Pathology, Iwate Medical University, Morioka; and 3Department
of Human Pathology, Gunma University Graduate School of Medicine, Maebashi, Japan

PPrimary granulomatous angiitis of the central nervous system (CNS) is extremely rare. Its preoperative diagnosis is
difficult as the condition displays nonspecific features on routine neuroimaging investigations. In this paper, the authors report findings of magnetic resonance (MR) spectroscopy and fractional anisotropy (FA) with diffusion tensor
MR imaging in a case of granulomatous angiitis of the CNS.
A 30-year-old man presented with morning headaches and grand mal seizures. An MR image revealed a mass resembling glioblastoma in the right temporal lobe. Magnetic resonance spectroscopy showed a high choline/creatine
(Cho/Cr) ratio indicative of a malignant neoplasm, accompanied by a slight elevation of glutamate and glutamine. The
FA value was very low, which is inconsistent with malignant glioma. The mass was totally removed surgically. Histologically, the peripheral lesion of the mass consisted of a rough accumulation of fat granule cells, infiltration of inflammatory cells, and distribution of capillary vessels. Some vessels within the lesion were replaced by granulomas. The
histological diagnosis was granulomatous angiitis of the CNS. The MIB-1–positive rate of the granuloma was approximately 5%. Both MR spectroscopy and FA were unable to accurately diagnose granulomatous angiitis of the CNS prior
to surgery; however, elevated Cho/Cr and glutamate and glutamine shown by MR spectroscopy may indicate the moderate proliferation potential of the granuloma and the inflammatory process, respectively, in this condition. Although
the low FA value in the present case enabled the authors to rule out a diagnosis of glioblastoma, FA values in inflammatory lesions require careful interpretation. (DOI: 10.3171/JNS-07/10/0873)
KEY WORDS • diffusion tensor imaging • fractional anisotropy •
magnetic resonance spectroscopy • MIB-1 • primary granulomatous angiitis • proliferation

P

RIMARY granulomatous angiitis of the CNS (also re-

ferred to as primary angiitis of the CNS or isolated
angiitis of the CNS) is an idiopathic inflammatory
disorder that solely targets small and middle parenchymal
and/or leptomeningeal vessels in the CNS.5,18,21 The diagnosis of granulomatous angiitis of the CNS must be reserved
for patients who have pathologically documented angiitis
with granulomatous features,4 whereas primary angiitis of
the CNS and isolated angiitis of the CNS emphasize the restricted nature of the vasculitis rather than the granulomatous histological nature.23 Granulomatous angiitis of the

Abbreviations used in this paper: Cho = choline; CNS = central
nervous system; Cr = creatine; DT = diffusion tensor; FA = fractional anisotropy; MR = magnetic resonance; NAA = N-acetylaspartate;
ROI = region of interest; VOI = voxel of interest.

J. Neurosurg. / Volume 107 / October, 2007

CNS is extremely rare; between its first mention in 19597
and 1997 only 136 cases had been reported in the literature.23 The origin, diagnosis, and natural course of granulomatous angiitis of the CNS remains unclear, even though it
has been nearly 50 years since the first report.
The diagnosis of granulomatous angiitis of the CNS is
commonly made based on specimens obtained at surgery,
and the disease responds effectively to corticosteroid and/or
cyclophosphamide treatment;23,24 however, the outcome is
fatal if patients are untreated because of delayed diagnosis.13,23 A preoperative diagnosis of granulomatous angiitis
of the CNS is essential for treatment; however, diagnosis is
difficult because neuroimaging findings, such as those on
computed tomography scanning, MR imaging, and cerebral
angiography are not specific.23 Magnetic resonance imaging
and angiography findings are not always in agreement.10
873

T. Beppu et al.
The MR images commonly reveal nonspecific abnormal
foci in the white matter, meningeal enhancement, and mass
lesions that are commonly misinterpreted as malignant neoplasms.12,19,23 Cerebral angiography reveals abnormal findings in 60% of patients, whereas “classic” findings of arteritis (alternating areas of stenosis and ectasia in multiple
vascular distributions) are observed in less than 40% of patients.3,23 Findings of other noninvasive neuroimaging examinations such as MR angiography, positron emission
tomography, and single-photon emission computed tomography are also not specific;18,23 consequently, an additional
examination is required that enables preoperative differential diagnosis for granulomatous angiitis of the CNS. In the
current study, we report findings of using single-voxel MR
spectroscopy and FA in DT imaging prior to surgery in a
case of granulomatous angiitis of the CNS.
Case Report
History and Examination. This 30-year-old man presented with morning headaches and grand mal seizures. On admission, neurological examination revealed no deficit other
than left homonymous upper quadrantanopia. His history
included no episode suggestive of systemic autoimmune
disease, intracranial infection, or head injury, and he had
not undergone previous surgery or radiation therapy. Physical examination revealed no skin abnormality. Results of
serum examinations, including a white blood cell count and
C-reactive protein test, were within the normal ranges. The
patient did not have temporal arteritis on admission. Computed tomography scanning and echography of the chest
failed to reveal any aortic arteritis, including Takayasu arteritis.
Neuroimaging. All image analyses including MR spectroscopy, FA for DT imaging, and routine MR imaging,
were performed using a 3.0-tesla MR imaging system (Signa VH/i, General Electric Medical Systems). Gadoliniumenhanced T1-weighted MR images revealed a mass with
peripheral enhancement resembling glioblastoma (Fig. 1).
In contrast, right carotid artery angiography revealed an
avascular area in the right temporal lobe and demonstrated
narrowing of the posterior temporal artery with poor filling
(Fig. 2). There was a contradiction between the MR imaging findings, which suggested a malignant neoplasm, and
those of angiography, which suggested a benign tumor or
any ischemic disease.
We performed single-voxel MR spectroscopy (TE 144
msec). The VOI was placed over the central and peripheral
regions of the mass lesion on T2-weighted MR images.
Magnetic resonance spectroscopy revealed remarkable elevations of Cho-containing compounds, Cr, and NAA in the
VOI at the peripheral region (Fig. 3), whereas the lipid peak
was observed in the central region (not shown in the figure). When we evaluated the MR spectroscopy findings following the final histological diagnosis, we considered that
the glutamate and glutamine peak was slightly elevated on
a spectroscopic pattern of the peripheral region (Fig. 3). We
interpreted the findings of MR spectroscopy for both regions prior to surgery and concluded that they were consistent with the pattern typical of an aggressive neoplasm such
as glioblastoma.
We also measured the FA value on DT imaging with a bfactor of 800 sec/mm2. We placed the ROI at the peripher874

FIG. 1. Axial Gd-enhanced T1-weighted MR image revealing
a peripheral enhancing mass resembling glioblastoma in the right
temporal lobe.

al enhancing and central necrotic regions of the lesion as
well as at the genu of the callosum on Gd-enhanced T1weighted images (Fig. 4). The FA value was highest in the
genu of the callosum (0.76); in contrast, FA values in the lesion were very low in both the peripheral enhancing region
(0.09) and in the central necrotic region (0.08). A marked-

FIG. 2. Right cerebral angiogram revealing upward displacement
of the main trunks of the middle cerebral artery due to a space-occupying lesion, in the absence of tumor blush vessels. Findings of narrowing and poor filling of the posterior temporal artery were also
observed (arrow).

J. Neurosurg. / Volume 107 / October, 2007

Primary granulomatous angiitis of the central nervous system

FIG. 4. Measurement of FA on DT imaging. The ROI was determined on a Gd-enhanced T1-weighted image (left) and transferred
onto an FA map (right). The FA values were then calculated using
modified Functool image analysis software (General Electric Medical Systems). Circles indicate ROIs within the genu of the splenium (1), the peripheral region (2), and the central necrotic region (3).

FIG. 3. Magnetic resonance spectroscopy findings. Upper: The
VOI (10 mm2) is placed at the peripheral regions of the mass on a
T2-weighted MR image (square). Lower: Graph showing marked
elevation of Cho at 3.2 ppm, Cr (Cre) at 3.0 ppm, and NAA at 2.0
ppm. A slight elevation of glucose and glutamate (Glx) is also observed at 2.1 to 2.2 ppm.

ly low FA value at the peripheral region enabled us to eliminate a diagnosis of a glioblastoma and speculate on the
possibility of an edematous and hypocellular lesion.
Operation. We performed a gross-total removal of the lesion because the lesion was causing considerable mass effect and widespread peripheral edema. When we opened the
dura mater, the cortical arachnoid and vessels appeared almost normal. The mass was very hard but not demarcated
from the surrounding xanthochromic edematous white matter. Postoperative MR imaging confirmed that the lesion
had been completely debulked.
J. Neurosurg. / Volume 107 / October, 2007

Histologically, the structure of the removed mass was revealed to consist of widespread necrotic tissue in the central
region with an inflammatory lesion in the peripheral region.
The inflammatory lesion consisted of a rough accumulation
of abundant fat granule cells and inflammatory cells such as
normal lymphocytes, astrocytes, and microglias, and included small hemorrhages, vessels infiltrated by lymphocytes, and vessels replaced by granuloma that consisted predominantly of proliferated epithelioid cells (Fig. 5A–C).
The histological diagnosis was granulomatous angiitis of
the CNS. The MIB-1 index within the granulomas was approximately 5% (Fig. 5D).
Postoperative Course. After surgery, treatment with lowdose (2 mg/day) bolus dexamethasone was initiated and
continued for 1 month, and the patient made a full recovery.
Three months after completion of the therapy, Gd-enhanced
T1-weighted images revealed multiple enhancing lesions in
the cerebral white matter bilaterally. The specimen obtained
from a stereotactic biopsy targeted to one of these lesions
showed similar histological features to those of the previous
lesion. The patient was treated initially with intravenous
high-dose dexamethasone (16 mg/day), which was tapered
gradually to 2 mg/day within 2 weeks; additionally, cyclophosphamide was administered as a bolus (100 mg/day).
A low-dose bolus of dexamethasone (2 mg/day) and bolus
of cyclophosphamide (100 mg/day) were continued for 3
months. The MR images obtained after the completion of
these therapies revealed the disappearance of all lesions.
The patient underwent rehabilitation therapy and currently
exhibits mild dementia.
Discussion
In the present case, we performed MR spectroscopy and
calculated the FA value by using DT imaging. In MR spectroscopy Cho and Cr were significantly elevated, suggesting
increased cell turnover and an energy-dependent system,
respectively. A high Cho/Cr ratio was also observed at the
periphery of the lesion. In the central region, the lipid peak
that indicates necrosis was remarkably elevated. We interpreted the findings of MR spectroscopy to represent the
high proliferation potential of a malignant neoplasm such as
glioblastoma. When observed retrospectively following his875

T. Beppu et al.

FIG. 5. Photomicrographs showing histological features of the inflammatory region of the mass. A: Low-power image
demonstrating that the lesion consists of a rough accumulation of inflammatory cells and includes small hemorrhages (H),
vessels infiltrated by lymphocytes (V), and vessels replaced by granuloma (G). B: A capillary vessel infiltrated by lymphocytes. C: A vessel partly occluded and replaced by granuloma consisting predominantly of proliferated epithelioid
cells. (Rough accumulations of fat granule cells and inflammatory cells are seen surrounding vessels in panels B and C.)
D: Demonstration of MIB-1–positive cells within the granuloma. H & E (A–C), original magnification 3 40 (A) and 3
200 (B–D).

tological diagnosis, we found a slightly elevated level of
glutamate and glutamine.
There has been only one report of MR spectroscopy performed in a patient with primary angiitis of the CNS in
which the authors described remarkable elevation of glutamate and glutamine in addition to the spectroscopic pattern
consistent with a neoplastic process.16 The authors proposed
that a marked elevation of glutamate and glutamine is associated with an inflammatory process of the CNS, and that
the observed elevation of glutamate and glutamine in primary angiitis of the CNS reflects the cell breakdown of
neural and glial elements, as well as the adjacent astrocytic
response, leading to the high concentration of glutamate
and glutamine in inflammatory conditions.
A remarkable elevation of glutamate and glutamine may
lead to preoperative diagnosis of any inflammatory disorder, including granulomatous angiitis of the CNS; however,
MR spectroscopy in the present case revealed a slight elevation of glucose and glutamine but marked elevations of
Cho and Cr. These findings possibly result from granulomas
retaining a moderate proliferation potential (MIB-1, ~ 5%).
To determine whether the granuloma in granulomatous
angiitis of the CNS generally retains a higher proliferation
potential, further reports of cumulatively increased MR
spectroscopy of granulomatous angiitis of the CNS are warranted.
The peripheral enhancing lesion on Gd-enhanced T1876

weighted images showed a low FA value of 0.09, whereas
the genu of the callosum displayed a high value of 0.76. The
callosum shows strong directionality of water diffusion and
consequently a high FA value.22 Previous reports of the FA
value of the callosum, 0.776 and 0.61,14 were similar to that
of the genu in the present case. This similarity confirms the
reliability of the FA values obtained in the present study.
The FA in the astrocytic tumor tissue is affected by a balance between the extent of nerve fiber destruction and an
increasing number of cells that represent spindle-shaped
neoplastic cells and elongated nuclei, and/or alignment of
cells in a preferred direction such as pseudopalisading.2,11,15,17
There is a tendency for tumors classified as high-grade gliomas to present with higher FA values.2,9 In previous studies, the mean FA values have been reported as 0.242 and
0.2311 in glioblastoma, and 0.23,11 0.19,2 and 0.179 in anaplastic astrocytomas. In the present case, the FA value for
the peripheral enhancing region was significantly lower
than those for glioblastoma and anaplastic astrocytoma. A
very low FA value enabled us to confidently propose prior to surgery that the peripheral lesion was different from
glioblastoma. Surgical specimens showed rough accumulations of fat granule cells and normal inflammatory cells in
most of the peripheral region, as shown in Fig. 5A to C. Extracellular edema and small hemorrhages were also observed. These histological characteristics may lead to a decrease in the degree of water diffusion directionality, and
J. Neurosurg. / Volume 107 / October, 2007

Primary granulomatous angiitis of the central nervous system
thereby to a largely reduced FA value; however, FA values
in inflammatory lesions are probably largely influenced by
the different degrees of inflammation among patients and
by the different histological characteristics among inflammatory disorders. For example, the FA values for multiple
sclerosis are different along the inflammatory process
(acute, subacute, and chronic plaques, and among regions of
the plaques).1,8,20 Although in the present case the FA was
helpful in the exclusive diagnosis of glioblastoma, FA values for inflammatory disorders require careful interpretation. Further study of FA in a limited group of patients with
angiitis of the CNS is required to confirm the suitability of
FA for diagnosis of angiitis, including granulomatous angiitis, of the CNS.
Conclusions
To our knowledge, there are no other reports of MR spectroscopy or calculations of the FA value for granulomatous
angiitis of the CNS. We emphasize that brain and meningeal
biopsy is the gold standard for the diagnosis of granulomatous angiitis of the CNS. Magnetic resonance spectroscopy
proved less useful for preoperative diagnosis for granulomatous angiitis of the CNS because it displayed a nonspecific spectroscopic pattern similar to that for malignant neoplasm, although a slight elevation of glucose and glutamine
indicative of inflammatory character was observed. However, these findings suggest the possibilities that: 1) granulomatous angiitis of the CNS retains moderate proliferation
potential; and 2) the elevation of the glucose and glutamate
peak is an indicator of any inflammatory disorder of the
CNS. In contrast, the low FA value in the present case was
sufficient to enable us to presume that the lesion was inconsistent with malignant gliomas; however, the use of FA to
interpret inflammatory disorders of the CNS should be predominantly influenced by differences in the histological behavior of inflammation among patients.
Acknowledgment
We thank Dr. Noriyuki Uesugi (Department of Pathology, Iwate
Medical University, Morioka, Japan) for preparation of the microphotographs.
References
1. Bammer R, Augustin M, Strasser-Fuchs S, Seifert T, Kapeller P,
Stollberger R, et al: Magnetic resonance diffusion tensor imaging for characterizing diffuse and focal white matter abnormalities in multiple sclerosis. Magn Reson Med 44:583–591, 2000
2. Beppu T, Inoue T, Shibata Y, Kurose A, Arai H, Ogasawara K,
et al: Measurement of fractional anisotropy using diffusion tensor MRI in supratentorial astrocytic tumors. J Neurooncol 63:
109–116, 2003
3. Calabrese LH: Vasculitis of the central nervous system. Rheum
Dis Clin North Am 21:1059–1076, 1995
4. Calabrese LH, Duna GF, Lie JT: Vasculitis in the central nervous system. Arthritis Rheum 40:1189–1201, 1997
5. Calabrese LH, Furlan AJ, Gragg LA, Ropos TJ: Primary angiitis of the central nervous system: diagnostic criteria and clinical approach. Cleve Clin J Med 59:293–306, 1992
6. Ciccarelli O, Werring DJ, Wheeler-Kingshott CA, Barker GJ,
Parker GJ, Thompson AJ, et al: Investigation of MS normal-appearing brain using diffusion tensor MRI with clinical correlations. Neurol 56:926–933, 2001

J. Neurosurg. / Volume 107 / October, 2007

7. Cravioto H, Feigin I: Noninfectious granulomatous angiitis with
a predilection for the nervous system. Neurol 9:599–609, 1959
8. Filippi M, Cercignani M, Inglese M, Horsfield MA, Comi G:
Diffusion tensor magnetic resonance imaging in multiple sclerosis. Neurology 56:304–311, 2001
9. Goebell E, Paustenbach S, Vaeterlein O, Ding XQ, Heese O,
Fiehler J, et al: Low-grade and anaplastic gliomas: differences in
architecture evaluated with diffusion-tensor MR imaging. Radiology 239:217–222, 2006
10. Greenan TJ, Grossman RI, Goldberg HI: Cerebral vasculitis: MR
imaging and angiographic correlation. Radiology 182:65–72,
1992
11. Inoue T, Ogasawara K, Beppu T, Ogawa A, Kabasawa H: Diffusion tensor imaging for preoperative evaluation of tumor grade
in gliomas. Clin Neurol Neurosurg 107:174–180, 2005
12. Ishikawa E, Tsuboi K, Takano S, Kimura H, Aoki T, Mashiko
R, et al: Primary cerebral angiitis containing marked xanthoma
cells with massive intraparenchymal involvement. Case report.
Neurol Med Chir (Tokyo) 45:156–160, 2005
13. Koo EH, Massey EW: Granulomatous angiitis of the central
nervous system: protean manifestations and response to treatment. J Neurol Neurosurg Psychiatry 51:1126–1133, 1988
14. Melhem ER, Itoh R, Jones L, Barker PB: Diffusion tensor
MR imaging of the brain: effect of diffusion weighting on trace
and anisotropy measurements. AJNR Am J Neuroradiol 21:
1813–1820, 2000
15. Misaki T, Beppu T, Inoue T, Ogasawara K, Ogawa A, Kabasawa H: Use of fractional anisotropy value by diffusion tensor
MRI for preoperative diagnosis of astrocytic tumors: case report.
J Neurooncol 70:343–348, 2004
16. Panchal NJ, Niku S, Imbesi SG: Lymphocytic vasculitis mimicking aggressive multifocal cerebral neoplasm: MR imaging
and MR spectroscopic appearance. AJNR Am J Neuroradiol
26:642–645, 2005
17. Sinha S, Bastin ME, Whittle IR, Wardlaw JM: Diffusion tensor
MR imaging of high-grade cerebral gliomas. AJNR Am J Neuroradiol 23:520–527, 2002
18. Siva A: Vasculitis of the nervous system. J Neurol 248:451–468,
2001
19. Tashiro J, Yokoyama N, Maruo Y, Kubota T, Niwa J, Shimoyama N: [A case of granulomatous angiitis of the central nervous
system presented with subacute mental deterioration resembling
diffuse white matter disease.] Rinsho Shinkeigaku 41:491–497,
2001 (Jpn)
20. Tievsky AL, Ptak T, Farkas J: Investigation of apparent diffusion
coefficient and diffusion tensor anisotropy in acute and chronic multiple sclerosis lesions. AJNR Am J Neuroradiol 20:
1491–1499, 1999
21. Vollmer TL, Guarnaccia J, Harrington W, Pacia SV, Petroff OA:
Idiopathic granulomatous angiitis of the central nervous system.
Diagnostic challenges. Arch Neurol 50:925–930, 1993
22. Witwer BP, Moftakhar R, Hasan KM, Deshmukh P, Haughton
V, Field A, et al: Diffusion-tensor imaging of white matter tracts
in patients with cerebral neoplasm. J Neurosurg 97:568–575,
2002
23. Younger DS, Calabrese LH, Hays AP: Granulomatous angiitis
of the nervous system. Neurol Clin 15:821–834, 1997
24. Younger DS, Hays AP, Brust JC, Rowland LP: Granulomatous
angiitis of the brain. An inflammatory reaction of diverse etiology. Arch Neurol 45:514–518, 1998
Manuscript submitted August 13, 2006.
Accepted February 28, 2007.
This work was supported by the Advanced Medical Science Center, Iwate Medical University, Japan.
Address correspondence to: Takaaki Beppu, M.D., Department of
Neurosurgery, Iwate Medical University, 19-1 Uchimaru, Morioka
020-8505, Japan. email: tbeppu@iwate-med.ac.jp.

877