CASE REPORT

Dural Arteriovenous Fistula Presents Like an Ischemic Stroke
Joanne R. Festa, PhD,* Ronald M. Lazar, PhD,*† Randolph S. Marshall, MD,*
John Pile-Spellman, MD,†‡§ Ji Y. Chong, MD,* and Hoang Duong, MD‡

Objective: We report the case of a patient with a dural arteriovenous
fistula whose neurobehavioral syndrome was indistinguishable from
that of an ischemic stroke.

Background: Case studies of dural arteriovenous fistulas primarily
describe global cognitive changes like dementia, but detailed neurocognitive evaluations of dural arteriovenous fistula patients are rarely
reported.
Method: We provide a dural arteriovenous fistula case of a patient
who presented with aphasia and other symptoms of stroke. Background history, serial neuropsychological data, and angiographic images are presented.
Results and Conclusions: Serial neurocognitive data show the
extent to which cognitive deficits are reversed with embolization. The
case demonstrates that the mechanisms underlying neurocognitive
deficits are specific to the fistula’s unique hemodynamic features in
addition to the location of the dural arteriovenous fistula.
Key Words: dural arteriovenous fistula, neurobehavioral manifestations, therapeutic embolization
(Cog Behav Neurol 2004;17:50–53)

D

ural arteriovenous fistulas (DAVF) are arteriovenous
shunts within the dural layer and account for approximately 1% of all strokes. DAVFs comprise 10 to 15% of cerebrovascular malformations, although some report that up to
20% of arteriovenous malformations (AVMs) are DAVFs.1
Treatments for DAVFs include neurovascular embolization,
radiosurgery, and surgical resection. Combined therapy and
staged embolization treatments are also often employed.
Received for publication May 30, 2003; revised August 25, 2003; accepted
September 23, 2003.
From the *Department of Neurology, Columbia University College of Physicians and Surgeons, New York, NY; †Department of Neurosurgery, Columbia University College of Physicians and Surgeons, New York, NY;
‡Department of Radiology, Columbia University College of Physicians
and Surgeons, New York, NY; and §Department of Anesthesiology, Columbia University College of Physicians and Surgeons, New York, NY.
Address correspondence to and reprint requests to Dr. Joanne Festa, Neurological Institute, 710 West 168th Street, New York, NY 10032, USA
(e-mail: jf2128@columbia.edu).
Copyright © 2004 by Lippincott Williams & Wilkins

50

DAVFs can mimic other neurologic disorders and produce a
broad spectrum of signs and symptoms, including headache,
tinnitus, vertigo, Parkinsonism, and visual field and gait disturbances.1–4 Studies of the neuropsychological sequelae of
DAVFs typically report global changes in cognitive status,
usually dementia.4–8 Detailed neurocognitive evaluations of
patients with DAVFs by standardized neuropsychological
tests, however, are rarely reported. We present a patient with a
DAVF followed with serial neuropsychological evaluations.
Findings suggest a pathophysiologic mechanism that may account for similar deficits also seen in brain AVMs.

CASE REPORT
An 80-year-old, right-handed woman with a 4-month
history of rambling in conversations experienced an episode of
severe confusion and was brought to an emergency room,
where she had a generalized seizure. She was found to have a
mild right hemiparesis that resolved within an hour. Brain imaging suggested a vascular malformation in the left temporooccipital region. She was placed on an anticonvulsant.
Two months after her initial presentation to the emergency room, she was referred for angiography and treatment
planning and underwent a neuropsychological assessment.9
Her educational history included a developmental reading
problem and 2 years of post high school course work. The neurocognitive evaluation revealed mild to moderate Wernicke
aphasia with greater impairment in auditory comprehension
than reading, short-term memory loss, and ideational apraxia
(Table 1, NE1). The patient was fully oriented on the Mini
Mental State Examination (MMSE),10 but could not spell
world backward, name a pencil, follow a 3-step command, recall 2 of 3 words after a brief delay, or copy intersecting pentagons. Her conversational speech was fluent without dysarthria or dysphonia, but the aphasia examination revealed literal
and verbal paraphasic errors, occasional perseveration, and a
disorder of comprehension. There was mild to moderate impairment in repetition, Right-Left orientation was extremely
slowed but accurate. Her writing mechanics suggested apraxia,
but her oral reading was intact. With extended time and many
self-corrections, she was able to draw the face of a clock with
the numbers correctly placed inside a circle, but she was unable to place hands indicating a designated time. Ideomotor
praxis was intact. Angiography demonstrated a left sigmoid
Cog Behav Neurol • Volume 17, Number 1, March 2004

Cog Behav Neurol • Volume 17, Number 1, March 2004

DAVF Presents Like Ischemic Stroke

TABLE 1. Neuropsychological Evaluation Test Scores (NE)

employing commands demonstrated some improvement, but
repetition remained impaired.
The patient underwent angiography 9 days after the initial treatment, targeting the distal left occipital artery for a second embolization. A superselective Wada test11 was performed that disclosed no evidence of visual field or cranial
nerve deficits, but the tortuosity of the targeted occipital artery
and the presence of vasospasm precluded embolization. Another neuropsychological examination 1 day later (10 days after the initial embolization) revealed continued improvement
in cognitive functioning (NE3). She showed a significant improvement in language and general cognitive functioning with
only a mild Wernicke aphasia. Her MMSE score was improved, and she had fewer paraphasias in conversation as well
as improvement in naming, comprehension, and repetition.
Angiography, intraoperative arterial exposure, and proximal
glue casting followed 6 weeks later. The distal branch feeders
of the left occipital artery and the left transmastoid branch
feeders were embolized using NBCA glue. Subsequent surgical resection sacrificed the remainder of the distal left occipital
artery. Postoperative angiography showed that arterial feeders
to the AVM were thrombosed and the DAVF was obliterated.
The following day, neuropsychological evaluation revealed
that the improvements in cognition seen previously were maintained (NE4).

Test9

NE 1

MMSE
R-L
Commands
CI
Repeating
AN
Oral
Reading
BN
Praxis
Calculations

*22/30
8/8
*11/15
*7/12
*11/16
*9
9/10
*7/10
*35/60
4/4
14/16

NE 2
(+2dy)

NE 3
(+10dy)

NE 4
(+6wk)

27/30

28/30
8/8
15/15
*6/12
14/16
11
10/10
9/10
42/60
4/4
15/16

27/30
8/8
14/15
*6/12
15/16
11
10/10
10/10
42/60

14/15
*4/8

*Indicates impaired performance. MMSE, Mini-Mental Status Examination; R-L, Right-left discrimination; CI, Complex ideational material; Repeating, Repeating phrases; AN, Animal naming; Oral, Oral sentence reading;
Reading, Reading sentences and paragraphs; BN, Boston Naming Test; P
raxis, Following 4 commands to gesture object use.

sinus DAVF supplied by bilateral occipital arteries and the
posterior division of the left middle meningeal artery. The left
sigmoid sinus was completely occluded, and there was retrograde drainage into cortical veins (Fig. 1). Embolization of the
left middle meningeal artery using Gelfoam and partial embolization of the right occipital feeder using NBCA glue were
performed successfully. Two days post embolization, her
MMSE improved to 27/30 (NE2). A test of comprehension

DISCUSSION
Our case demonstrates that the neurologic presentation
of dural arteriovenous fistulas can mimic focal neurologic disorders of the brain. Although the main lesion was outside of the

FIGURE 1. Angiogram showing venous congestion caused by sinus occlusions (occluded unopacified transverse sinus, A, dotted
line indicates sinus) and concomitant dural fistula. The congestion is centered over the posterior temporal lobe with retrograde
venous drainage from isolated sinus segment (isolated segment of transverse sinus, B), back through the vein of Labbe C), into
the middle cerebral vein (D), and into vein of Trolard (E). Additional drainage of congested temporal lobe retrograde to the
occipital temporal vein (F). Arterialized blood supplied to the fistula (fistula rete *) via shunts fed by the middle meningeal artery
(G) and the posterior auricular artery (H).
© 2004 Lippincott Williams & Wilkins

51

Festa et al

brain parenchyma, this patient’s Wernicke aphasia, transient
right hemiparesis, confusion, memory loss, and ideational
apraxia were indistinguishable from symptoms found in patients with focal brain lesions from stroke, brain tumors, or
infection. On presentation to the emergency room, the rapid
resolution of her symptoms raised the possibility that the deficits were postictal. The patient’s family reported that prior to
her emergency room presentation, she was rambling in conversation. This is suggestive of aphasia, but the etiology is unclear. However, 2 months after her hospitalization, the neurocognitive examination demonstrated a persistent aphasia,
making it unlikely that her syndrome was due to a postictal
state.
The spectrum of signs and symptoms associated with
DAVFs has been attributed to any one or combination of 6
pathophysiologic factors: the arteriovenous shunt, cerebral
hypoxia and ischemia, increased venous pressure, retrograde
drainage and secondary engorgement of basal sinuses, sinus
obstruction, and subarachnoid hemorrhage secondary to involvement of the pial venous system.12 Passive venous hypertension and/or congestion is often cited as an important etiological factor in the development of deficits in DAVFs.5,6,13–16
In one series, 20% of the DAVF patients with persistent cortical venous reflux suffered a progressive dementia syndrome
due to generalized venous hypertension.13 Venous hypertension, due to increased inflow, reduction of outflow, or both,
can lead to retrograde transmission of pressure into the cortical
veins, which in turn impairs parenchymal venous drainage,
thereby causing ischemia. Reduced regional cerebral blood
flow and increased regional oxygen extraction fraction have
been associated with cognitive symptoms, with the impaired
hemodynamics presumed to arise from the venous congestion
at the site corresponding to the symptoms.17 Parenchymal circulation delay was also seen in areas with reduced regional
cerebral blood flow.
In a review of 100 cases, Awad et al found that DAVF
location determined the type of symptoms, but drainage
characteristics were responsible for the focal neurologic
complications.18 Lasjaunias et al also related focal central nervous system symptoms to the venous territory and not to the
arterial supply characteristics.14 Focal neurologic deficits in
cases of nonhemorrhagic brain AVMs have also been postulated to be the result of venous hypertension rather than “cerebral steal,” as has been argued in the past.19 In the context
of DAVFs, meningeal arterial steal has been associated with
cranial nerve deficits due to the dural arteries’ supply to the
cranial nerves.14 However, since the intracranial internal
carotid and vertebrobasilar systems supply the brain, but
contribute little to the dura, arterial steal is no longer considered a tenable explanation for symptoms associated with
DAVFs.2,7,14,20
This case illustrates the effects of embolization on neurocognitive functioning with longitudinal psychometric data.

52

Cog Behav Neurol • Volume 17, Number 1, March 2004

The DAVF characteristics responsible for the observed neurologic deficits in this patient, we believe, were the bilateral arterial supply from 3 arterial feeders combined with the complete occlusion of the left sigmoid sinus causing significant
venous backpressure into the cerebral cortex. This mechanism
of venous congestion and consequent hypoperfusion is a recognized mechanism for neurologic deficits found in brain
AVMs.21 Global venous hypertension may present with nonspecific symptoms such as headache and reduced level of consciousness. When the venous pressure is transmitted to a more
restricted region, focal symptoms can occur. Ischemic lesions
due to arterial occlusions may be indistinguishable clinically
from those due to venous occlusion. Arterial ischemic lesions,
however, appear on imaging in a specific arterial territory,
whereas venous ischemic lesions often cross arterial vascular territories and have associated edema. Arteriography is often
needed, as in this case, to diagnose the underlying cause of the
ischemia.
In this patient, the presumed reduction of pressure from
the arterial side with the 2 arterial embolizations lowered the
arterial inflow into the DAVF, resulting in an alleviation of
neurologic symptoms. The cognitive deficits and resulting improvements in cognition, however, were associated with areas
not adjacent to the embolized arteries but rather in regions distant from the treatment site. A possible mechanism for the neurologic recovery was the normalization of local hemodynamics by embolization in the occipital region and the subsequent
reduction of parenchymal ischemia due to reduced venous
backpressure in the left temporal lobe.
Various interrelated mechanisms may contribute to the
presenting syndromes in any particular DAVF. The mechanisms associated with a patient’s symptom profile are specific
to the fistula’s hemodynamic features and the location of that
DAVF, with a behavioral syndrome that may be indistinguishable from that arising from an ischemic stroke.
REFERENCES
1. Kurl S, Saair T, Vanninen R, et al. Dural arteriovenous fistulas of superior
sagittal sinus: Case report and review of the literature. Surg Neurol. 1996;
45:250–255.
2. Hirono N, Yamadori A, Komiyama M. Dural arteriovenous fistula: A
cause of hypoperfusion-induced intellectual impairment. Eur Neurol.
1993;33:5–8.
3. Vinuela F, Fox AJ, Pelz DM, et al. Unusual clinical manifestations of
dural arteriovenous malformations. J Neurosurg. 1986;64:554–558.
4. Matsuda S, Wargai M, Shinotoh H, et al. Intracranial dural arteriovenous
fistula (DAVF) presenting progressive dementia and parkinsonism. J
Neurol Sci. 1999;165:43–47.
5. Hurst RW, Bagley LJ, Galetta S, et al. Dementia resulting from dural
arteriovenous fistulas: The pathologic findings of venous hypertensive
encephalopathy. Am J Neuroradiol. 1998;19:1267–1273.
6. Zeidman SM, Monsein LH, Arosarena O, et al. Reversibility of white
matter changes and dementia after treatment of dural fistulas. Am J Neuroradiol. 1995;16:1080–1083.
7. Nencini P, Inzitari D, Gibbs J, et al. Dementia with leucoaraiosis and dural
arteriovenous malformation: clinical and PET case study. J Neurol Neurosurg Psychiatry. 1993;56:929–931.
8. Ito M, Sonokawa T, Mishina H, et al. Reversible dural arteriovenous mal-

© 2004 Lippincott Williams & Wilkins

Cog Behav Neurol • Volume 17, Number 1, March 2004

formation-induced venous ischemia as a cause of dementia: Treatment by
surgical occlusion of draining dural sinus: case report. Neurosurgery.
1995;37:1187–1192.
9. Lezak MD. Neuropsychological Assessment. 3rd ed. New York, NY: Oxford University Press; 1995.
10. Folstein MF, Folstein SE, McHugh PR. Mini-mental state. J Psychiatr
Res. 1975;12:189–198.
11. Lazar RM, Marshall RS, Pile-Spellman J, et al. Anterior translocation of
language in patients with left cerebral arteriovenous malformation. Neurology. 1997;49:802–808.
12. Obrador S, Soto M, Silvela J. Clinical syndromes of arteriovenous malformations of the transverse-sigmoid sinus. J Neurol Neurosurg Psychiatry. 1975;38:436–451.
13. van Dijk JMC, terBrugge KG, Willinsky RA, et al. Clinical course of
dural arteriovenous fistulas with long-term persistent cortical venous reflux. Stroke. 2002;33:1233–1236.
14. Lasjaunias P, Chiu M, Brugge KR, et al. Neurological manifestations of
intracranial dural arteriovenous malformations. J Neurosurg. 1986;64:
724–730.

© 2004 Lippincott Williams & Wilkins

DAVF Presents Like Ischemic Stroke

15. Tomlinson FH, Rufenacht DA, Sundt TM, et al. Arteriovenous fistulas of
the brain and the spinal cord. J Neurosurg. 1993;79:16–27.
16. Cognard C, Casasco A, Toevi M, et al. Dural arteriovenous fistulas as a
cause of intracranial hypertension due to impairment of cranial venous
outflow. J Neurol Neurosurg Psychiatry. 1998;65:308–316.
17. Iwama T, Hashimoto N, Takagi Y, et al. Hemodynamic and metabolic
disturbances in patients with intracranial dural arteriovenous fistulas: positron emission tomography evaluation before and after treatment. J Neurosurg. 1997;86:806–811.
18. Awad IA, Little JR, Akrawi WP, et al. Intracranial dural arteriovenous
malformations: factors predisposing to an aggressive neurologic course. J
Neurosurg. 1990;72:839–850.
19. Mast J, Mohr JP, Osipov A, et al. Steal” is an unestablished mechanism for
the clinical presentation of cerebral arteriovenous malformations. Stroke.
1995;26:1215–1220.
20. Deveikis JV. Venous hypertensive encephalopathy. Am J Neuroradiol.
1998;19:1274–1275.
21. The Arteriovenous Malformation Study Group. Arteriovenous malformations of the brain in adults. N Engl J Med. 1999;340:1813–1818.

53