OPTIC NERVE ARTERIOVENOUS MALFORMATION
CAUSING OPTIC APOPLEXY: CASE REPORT
Yoshihiro Muragaki, M.D.
Department of Neurosurgery,
Neurological Institute, Tokyo
Women’s Medical University,
Tokyo, Japan

Hiroshi Ujiie, M.D.
Department of Neurosurgery,
Neurological Institute, Tokyo
Women’s Medical University,
Tokyo, Japan

Masayuki Ohno, M.D.
Department of Neurosurgery,
Neurological Institute, Tokyo
Women’s Medical University,
Tokyo, Japan

Osami Kubo, M.D.
Department of Neurosurgery,
Neurological Institute, Tokyo
Women’s Medical University,
Tokyo, Japan

Tomokatsu Hori, M.D.
Department of Neurosurgery,
Neurological Institute, Tokyo
Women’s Medical University,
Tokyo, Japan
Reprint requests:
Yoshihiro Muragaki, M.D.,
Department of Neurosurgery,
Neurological Institute, Tokyo
Women’s Medical University, 8-1
Kawada-cho, Shinjuku-ku, Tokyo
162-8666, Japan.
Email:
ymuragaki@nij.twmu.ac.jp
Received, September 5, 2001.
Accepted, February 20, 2002.

NEUROSURGERY

OBJECTIVE AND IMPORTANCE: Vascular malformations in the optic pathway are
rare. Only one case of pathologically confirmed arteriovenous malformation (AVM) of
the optic nerve has been reported previously. We document the case of a patient with
an optic nerve AVM who presented with optic apoplexy that was diagnosed with the
use of magnetic resonance imaging.
CLINICAL PRESENTATION: A 15-year-old girl developed left visual disturbance of
sudden onset while playing badminton. A magnetic resonance imaging scan disclosed
left optic nerve swelling and intraoptical hemorrhage, although an angiogram did not
reveal abnormal vessels.
INTERVENTION: The patient underwent total removal of the hematoma and tangles of the
abnormal vessels in the left optic nerve, which was diagnosed pathologically as an AVM. The
patient recovered visual acuity, but the left visual field defect remained unchanged.
CONCLUSION: Along with cavernous malformations and optic gliomas, AVMs can be a
rare cause of optic nerve apoplexy. T2-weighted magnetic resonance imaging is useful in
rendering the diagnosis of an optic nerve AVM, observed as a mass lesion consisting of
serpiginous, tangled, low-intensity bands. Early surgical treatment is recommended to
obtain a rapid recovery.
KEY WORDS: Angiographically occult vascular malformation, Optic nerve apoplexy, Thrombosed
arteriovenous malformation
Neurosurgery 51:1075-1078, 2002

DOI: 10.1227/01.NEU.0000027882.04518.98

A

rteriovenous malformations (AVMs)
are common vascular lesions in the
brain parenchyma. AVMs in the cranial
nerves, however, are quite rare (8, 9, 16, 18, 21,
23). We present one case of a patient with an
AVM in the optic nerve. We review the clinical presentation; radiological characteristics,
including magnetic resonance imaging (MRI)
scans; and the pertinent literature.

CASE REPORT
A previously healthy 15-year-old girl experienced the sudden onset of left visual acuity
disturbance while playing badminton on January 11, 1999. Because of the worsening
course of her visual acuity, her ophthalmologist obtained a computed tomographic scan,
which demonstrated the presence of a suprasellar mass. As a result, the patient was referred to our department on January 28, 1999.
At the first clinical examination, the patient’s visual acuity was 1.2 in the right eye
without any visual field defect and counting
fingers in the left eye. Intraocular pressure,
eye movement, and color vision all were normal. An ophthalmoscopic examination revealed normal findings in both eyes.
A cranial x-ray revealed no enlargement of
either the bilateral optic canals or the sella tur-

www.neurosurgery-online.com

cica. A computed tomographic scan revealed a
hyperdense suprasellar mass with slight calcification and no enhancement (Fig. 1). The tentative diagnosis at admission was pituitary apoplexy. An MRI scan was obtained 6 days after
admission. The T1-weighted MRI scan, especially the sagittal view parallel to the left optic
nerve (oblique sagittal view), revealed that the
swelling of the left intracranial optic nerve consisted of a mixed-intensity mass, which was
slightly enhanced after the administration of
gadolinium (Fig. 2). The T2-weighted scan demonstrated an abnormal, serpiginous, lowintensity mass surrounded by a high-intensity
area in the left optic nerve (Fig. 2). Carotid angiograms demonstrated elevation of the anterior
cerebral arteries; however, no abnormal vessel
was detected. These findings suggested an angiographically occult vascular malformation
(AOVM) with hematoma. We suspected tumor
bleeding of optic glioma preoperatively in this
patient because AOVM in the cranial nerves is
rare. The patient also described a temporal visual field defect in her right eye just before
surgery.
The patient underwent surgery 8 days after
admission (i.e., 26 days after the onset of symptoms). A bifrontal craniotomy was performed
and an interhemispheric approach revealed the
swelling of the left optic nerve with an old he-

VOLUME 51 | NUMBER 4 | OCTOBER 2002 | 1075

MURAGAKI ET AL.

matoma (Fig. 3). The medial
inferior pia of the left optic
nerve exhibited a hematoma,
and tortuous small abnormal
vessels were observed within
the hematoma. The hematoma and abnormal vessels
were totally removed pieceby-piece. Although the superior hypophyseal artery from
the left internal carotid artery
seemed to be the feeder of the
AVM, no definite drainer
was detected.
A pathological examina- FIGURE 1. Plain computed tomographic scan showing a suprasellar,
tion confirmed the diagnosis
isodense mass lesion including small,
of AVM in the optic nerve. spotty calcification.
The observed vascular malformations contained both
arterial and venous components, which were confirmed by use
of elastica van Gieson stain, and intervening glial tissue also was
identified by immunostaining with antiglial fibrillary acidic protein antibody (Fig. 4). In addition, extravasated red blood cells (a
sign of recent hemorrhage), hemosiderin-laden macrophages (a
sign of old hemorrhage), thrombosed malformed vessels, and
calcified vessel walls all were observed.
The surgical outcome was favorable, and clinical improvement was observed immediately after the operation. At discharge, the patient’s visual acuity had recovered to 0.6 in the
left eye, with temporal hemianopsia, and the symptoms of
upper temporal quadrantanopia were improving. An MRI
scan obtained 1 year after surgery revealed complete disappearance of abnormal mass in the optic nerve. At 15 months of
follow-up, the patient was doing well, and her visual acuity
remained unchanged.

DISCUSSION
McCormick (13) described different well-distinguished pathological entities, including cavernous malformations (CMs), capillary angiomas, venous malformations, and AVMs. Vascular
malformations in the optic pathway are rare, however. To date,
24 cases have been reported in the literature, including 17 CMs
(1–6, 10, 11, 14, 15, 20, 22, 26, 27) and 7 AVMs (8, 9, 16, 18). Five
of the reported AVMs were localized in the chiasm, and two
AVMs were in the optic nerve; however, only one case of pathologically confirmed AVM in the optic nerve has been reported
(16). AVMs that develop only in the cranial nerves also are highly
uncommon. As far as we know, only two previous reports have
described AVMs intrinsic to cranial nerves other than the optic
nerve: Tsubaki et al. (23) reported two cases of impaired trigeminal nerve, and Sindou et al. (21) reported a case of an impaired
trochlear nerve.
As in this patient, onset of vascular malformation in the
optic pathway tends to demonstrate a sudden disturbance in
both visual acuity and the visual field. Reilly and Oatey (16)

1076 | VOLUME 51 | NUMBER 4 | OCTOBER 2002

FIGURE 2. MRI scans. A, T1weighted image showing the suprasellar, mixed-signal-intensity lesion.
Note the isointensity consistent with
tissue matrix and multiple round
high signal intensity consistent with
subacute blood (methemoglobin). B,
T2-weighted image showing serpiginous, tangled low intensity suggestive of abnormal vessels. C,
T1-weighted oblique sagittal image
parallel to the left optic nerve clearly showing abnormal mixed intensity in the
optic nerve.

used the term optic nerve apoplexy for the first time in
two cases of hemorrhage in
the optic nerve caused by
AVMs. Maitland et al. (9)
also described four cases of
“chiasmal
apoplexy”
caused by intrachiasmal hematomas (three caused by
cryptic AVMs and one secondary to a glioma). Our
patient also developed optic
nerve apoplexy at onset,
which was followed by
abrupt worsening of the vi- FIGURE 3. Operative photograph
sual field. A visual defect of showing a subpial, firm, black mass
the upper quadrant in the distending the medial aspect of the
contralateral eye is consid- left optic nerve.
ered the result of minor rebleeding or edema.
All reported optochiasmal vascular malformations in the
literature had angiographically occult features (i.e., AOVMs)
and such AOVMs were not visualized angiographically (1–6,
8–11, 14–18, 20, 22, 26, 27). The different pathological subtype
of AOVM did not exhibit clearly different macroscopic findings. Lobato et al. (7) demonstrated that 59 (52%) of 114
angiographically occult AVMs and that 41 (50%) of 82 CMs
occurred with brain hematoma. In addition, they demonstrated that 13 (11%) of 114 angiographically occult AVMs and
12 (15%) of 82 CMs occurred with subarachnoid hemorrhage.
MRI quickly has become the primary modality for rendering
accurate differential diagnoses of AOVMs and other diseases.

www.neurosurgery-online.com

OPTIC NERVE ARTERIOVENOUS MALFORMATION

intervening brain tissue. In contrast, the diagnosis of AVM
was made on the basis of the presence of a compact bundle of
vessels that at least partially contained elastic lamina and
smooth muscle, intervening brain tissue, and vessels of varying thickness and hyalinization. Abnormal vessels usually are
identified with some degree of thrombosis. In the series of
histologically confirmed occult AVMs reported by Wakai et al.
(25), all patients had partially thrombosed vessels and evidence of hemorrhage. Microscopic findings in our patient
revealed various sizes of abnormal vessels containing an elastic lamina and smooth muscle, as confirmed with elastica van
Gieson staining (Fig. 4), as well as intervening gliotic tissue, as
confirmed with glial fibrillary acidic protein immunostaining
(Fig. 4). Thus, the microscopic findings (Fig. 4), together with
the MRI findings (Fig. 2), strongly suggest that the patient
revealed typical features of AVM but not CM.
Surgical treatment is justified for diagnostic purposes, decompression of the optic nerve, and prevention of lesion
growth and recurrent hemorrhage (16). Our patient presented
with an abrupt worsening of visual field defect during the
clinical course, which suggested rebleeding or a circulatory
disorder caused by mass effect. Although Lavin et al. (6)
reported a patient exhibiting symptom recovery without surgical maneuver, the present patient emphasizes that an early
extirpation is recommended to achieve an improvement in
symptoms of blurred vision and also to prevent a further
deterioration.

REFERENCES
FIGURE 4. Photomicrographs of AVM specimens. A, specimen showing
various size of arteries and veins (hematoxylin and eosin; original magnification, ⫻40). B, specimen showing abnormal vessels containing arterial
components with elastic lamina and venous components (elastica van Gieson; original magnification, ⫻40). C, specimen showing intervening glial
tissue among abnormal vasculature (antiglial fibrillary acidic protein antibody; original magnification, ⫻40).

Vanefsky et al. (24) reported differences in the MRI findings of
each AOVM. According to their findings, lesions that exhibit a
ring of hemosiderin are associated with CM components. In
contrast, the characteristics of AVM components include surrounding edema, mass effect, a single prominent blood product, and an expansile hemorrhage.
Our report is the first description of the MRI findings of an
optic nerve AVM. The MRI scans of our patient revealed mass
effect and an expansile hemorrhage, and these findings are
compatible with those for AVMs. In addition, no hemosiderin
rim was observed. An oblique sagittal view parallel to the
optic nerve is very useful, not only to distinguish optic nerve
apoplexy from pituitary apoplexy but also to assess these
characteristics of the MRI findings. However, the definitive
diagnosis of an AOVM depends on its pathological diagnosis
(12, 13, 17, 19). The diagnosis of CM was made when the
specimen exhibited back-to-back, thin-walled sinusoids that
were devoid of elastic lamina and smooth muscle, with no

NEUROSURGERY

1. Castel JP, Delorge-Kerdiles C, Rivel J: Cavernous angioma of the optic
chiasm [in French]. Neurochirurgie 35:252–256, 1989.
2. Corboy JR, Galetta SL: Familial cavernous angiomas manifesting with an
acute chiasmal syndrome. Am J Ophthalmol 108:245–250, 1989.
3. Hassler W, Zentner J, Petersen D: Cavernous angioma of the optic nerve:
Case report. Surg Neurol 31:444–447, 1989.
4. Iwai Y, Yamanaka K, Nakajima H, Miyaura T: Cavernous angioma of the
optic chiasm: Case report. Neurol Med Chir (Tokyo) 39:617–620, 1999.
5. Klein LH, Fermaglich J, Kattah J, Luessenhop AJ: Cavernous hemangioma of
optic chiasm, optic nerves and right optic tract: Case report and review of
literature. Virchows Arch A Pathol Anat Histol 383:225–231, 1979.
6. Lavin PJ, McCrary JA III, Roessmann U, Ellenberger C Jr: Chiasmal apoplexy: Hemorrhage from a cryptic vascular malformation in the optic chiasm. Neurology 34:1007–1011, 1984.
8. Lejeune JP, Hladky JP, Dupard T, Parent M, Hache JC, Christiaens JL:
Opticochiasmatic apoplexy [in French]. Neurochirurgie 36:303–307, 1990.
7. Lobato RD, Perez C, Rivas JJ, Cordobes F: Clinical, radiological, and pathological spectrum of angiographically occult intracranial vascular malformations: Analysis of 21 cases and review of the literature. J Neurosurg 68:518–
531, 1988.
9. Maitland CG, Abiko S, Hoyt WF, Wilson CB, Okamura T: Chiasmal apoplexy: Report of four cases. J Neurosurg 56:118–122, 1982.
10. Malik S, Cohen BH, Robinson J, Fried A, Sila CA: Progressive vision loss: A
rare manifestation of familial cavernous angiomas. Arch Neurol 49:170–173,
1992.
11. Maruoka N, Yamakawa Y, Shimauchi M: Cavernous hemangioma of the
optic nerve: Case report. J Neurosurg 69:292–294, 1988.
12. Matias-Guiu X, Alejo M, Sole T, Ferrer I, Noboa R, Bartumeus F: Cavernous
angiomas of the cranial nerves: Report of two cases. J Neurosurg 73:620–622,
1990.

VOLUME 51 | NUMBER 4 | OCTOBER 2002 | 1077

MURAGAKI ET AL.

13. McCormick WF: Pathology of vascular malformations of the brain, in Wilson CB, Stein BM (eds): Intracranial Arteriovenous Malformations. Baltimore,
Williams & Wilkins, 1984, pp 44–63.
14. Mohr G, Hardy J, Gauvin P: Chiasmal apoplexy due to ruptured cavernous
hemangioma of the optic chiasm. Surg Neurol 24:636–640, 1985.
15. Regli L, de Tribolet N, Regli F, Bogousslavsky J: Chiasmal apoplexy: Haemorrhage from a cavernous malformation in the optic chiasm. J Neurol
Neurosurg Psychiatry 52:1095–1099, 1989.
16. Reilly PL, Oatey PE: Optic nerve apoplexy: Report of two cases. J Neurosurg
64:313–316, 1986.
17. Robinson JR Jr, Awad IA, Masaryk TJ, Estes ML: Pathological heterogeneity
of angiographically occult vascular malformations of the brain. Neurosurgery 33:547–555, 1993.
18. Roski RA, Gardner JH, Spetzler RF: Intrachiasmatic arteriovenous malformation: Case report. J Neurosurg 54:540–541, 1981.
19. Russell D, Rubinstein L: Pathology of Tumors of the Nervous System. Baltimore,
Williams & Wilkins, 1977, ed 4, pp 116–145.
20. Shibuya M, Baskaya MK, Saito K, Suzuki Y, Ooka K, Hara M: Cavernous
malformations of the optic chiasma. Acta Neurochir (Wien) 136:29–36, 1995.
21. Sindou M, Gilg A, Vighetto A, Jouvet A: Cryptic angioma in the trochlear
nerve: Excision of the invaded portion and successful repair with an autologous graft—Case report. Neurosurgery 30:255–258, 1992.
22. Steinberg GK, Marks MP, Shuer LM, Sogg RL, Enzmann DR, Silverberg GD:
Occult vascular malformations of the optic chiasm: Magnetic resonance
imaging diagnosis and surgical laser resection. Neurosurgery 27:466–470,
1990.
23. Tsubaki S, Fukushima T, Tamagawa T, Miyazaki S, Watanabe K, Kuwana N,
Shimizu T: Parapontine trigeminal cryptic angiomas presenting as trigeminal neuralgia. J Neurosurg 71:368–374, 1989.
24. Vanefsky MA, Cheng ML, Chang SD, Norbash A, Snipe J, Marks MP,
Steinberg GK: Correlation of magnetic resonance characteristics and histopathological type of angiographically occult vascular malformations.
Neurosurgery 44:1174–1181, 1999.
25. Wakai S, Ueda Y, Inoh S, Nagai M: Angiographically occult angiomas: A
report of thirteen cases with analysis of the cases documented in the literature. Neurosurgery 17:549–556, 1985.
26. Warner JE, Rizzo JF III, Brown EW, Ogilvy CS: Recurrent chiasmal apoplexy
due to cavernous malformation. J Neuroophthalmol 16:99–106, 1996.
27. Zentner J, Grodd W, Hassler W: Cavernous angioma of the optic tract.
J Neurol 236:117–119, 1989.

AVM. Still, it is difficult to explain how an optic nerve AVM
could rupture without producing subarachnoid hemorrhage
and to reconcile the presence of patent arteries histologically
with a negative angiogram. One is left with a troubling discrepancy between the clinical and pathological diagnoses, and
to assign the diagnosis of angiographically occult vascular
malformation is not a satisfying way to settle the issue. Therefore, this report reminds neurosurgeons to carefully distinguish between cavernous malformations and AVMs and to
reserve the diagnosis of an angiographically occult vascular
malformation for this unresolved clinical and pathological
inconsistency.
Michael T. Lawton
San Francisco, California

T

his interesting case report describes a patient with an
angiographically occult AVM in the optic nerve with hemorrhage and acute unilateral visual deterioration. Microsurgical resection of the cryptic AVM and hematoma resulted in
improved visual acuity and improved right eye visual field
defect, from a temporal hemianopsia to a temporal upper
quadrantanopia. Although vascular malformations of the optic pathways have been reported previously, few of these have
been true AVMs, and only two occurred in the optic nerve. In
this patient, I do not think that the magnetic resonance imaging features of the vascular malformation and hemorrhage can
distinguish an angiographically occult AVM from a cavernous
malformation. The treatment would be similar, and I agree
with the choice of microsurgical resection of the vascular
lesion, taking meticulous care to dissect along the edge of the
malformation to avoid injury to the normal optic nerve. The
authors achieved an excellent clinical outcome with surgery.
Gary K. Steinberg
Stanford, California

COMMENTS

A

case of a 15-year-old girl who presented with sudden
vision loss in the left eye is described as an acutely ruptured arteriovenous malformation (AVM). However, the computed tomographic and magnetic resonance imaging scans
have an appearance characteristic of cavernous malformations, and the angiogram revealed no arteriovenous shunting,
nidus, or draining vein. At surgery, a small tangle of vessels
was found, without a discrete draining vein. The diagnosis of
AVM was made after pathological analysis. This case illustrates the “gray area” created by the diagnosis termed angiographically occult vascular malformation. On the basis of its radiographic and gross appearance, most radiologists and
neurosurgeons would be quick to call this lesion a cavernous
malformation. An intracapsular hemorrhage could easily
present without subarachnoid hemorrhage, as did this one. In
contrast, on the basis of its histological appearance, most
pathologists would identify features more consistent with

T

he authors’ recommendation that a patient such as the one
reported here must undergo surgery seems wise, not only
because clinical history and imaging were compatible with the
diagnosis of hemorrhagic tumor but also because the patient
could benefit from nerve decompression and resection of the
lesion. The authors provide a review of the recent literature on
the subject, which will be useful for the neurosurgical community. My sole concern is the AVM terminology used in the
title. Because no abnormal vessels were detected with angiography, angiographically occult AVMs or angiographically occult
vascular malformations would be better terms to use, despite the
demonstration of the existence among abnormal vasculature
of a cluster of arteries and veins as well as glial tissue, which
are compatible with telangiectasias.
Marc P. Sindou
Lyon, France