De Novo Cerebral Arteriovenous Malformation

De Novo Cerebral Arteriovenous
Malformation: Case Report
Ketan R. Bulsara, M.D., Michael J. Alexander, M.D.,
Alan T. Villavicencio, M.D., Carmelo Graffagnino, M.D.
Divisions of Neurosurgery (KRB, MJA, ATV) and Neurology (CG), Duke University Medical
Center, Durham, North Carolina

OBJECTIVE AND IMPORTANCE: Arteriovenous malformations (AVMs) are
generally thought to have a congenital cause. This is the first report of an
angiographically proven de novo cerebral AVM in an adult patient without previous vascular abnormality.
CLINICAL PRESENTATION: A 26-year-old African-American woman developed multiple cranial nerve deficits and ataxia over the course of a few
days after a streptococcal throat infection. T2-weighted magnetic resonance imaging scans revealed a hyperintense signal in the midbrain with
extension into the diencephalon. A cerebral angiogram performed at that
time to exclude vasculitis revealed normal cerebral vasculature. The
patient was treated with corticosteroids, and symptoms resolved. Subsequently, at the age of 32, this patient presented with a severe headache
and emesis, but with no focal neurological deficit.
INTERVENTION: The patient’s cranial computed tomographic and magnetic
resonance imaging scans revealed a right posterior temporal intraparenchymal
hemorrhage, and cerebral angiography revealed a new 3- by 2-cm AVM. The
patient underwent microsurgical resection of the AVM and associated hematoma. Postoperative angiography revealed no evidence of residual AVM.
CONCLUSION: This study details the case of a woman who developed a de
novo cerebral AVM during a 6-year period. This report challenges the
conventional belief that all AVMs have a congenital cause.
(Neurosurgery 50:1137–1141, 2002)
Key words: Acquired arteriovenous malformation, Angiogenesis, Cerebral arteriovenous malformation

I

ntracranial arteriovenous malformations (AVMs), by definition, anatomically lack a capillary bed but demonstrate an arterial nidus, distinguishing
them from pure arteriovenous fistulae
(20). AVMs, however, may incorporate
high-flow
arteriovenous
shunting
through one or more direct fistulae. The
annual risk of hemorrhage in cerebral
AVMs is 2 to 4% (31), but it can be as
high as 18% with previous hemorrhage
(19). The accepted congenital cause of
AVMs has been questioned (26, 27). Although it is conceivable that some

AVMs have an underlying developmental derangement, many correlations suggest that this may not be true for all of
them (17, 21, 27). To our knowledge, this
is the first report of an angiographically
proven de novo intracranial AVM in an
adult patient without previous vascular
abnormality.

CASE REPORT
A 26-year-old African-American
woman developed intermittent monocular vision loss and progressive ataxia
Neurosurgery, Vol. 50, No. 5, May 2002

1137

several days after developing a severe
streptococcal throat infection. The patient used amphetamines chronically for
weight reduction. Two weeks after her
initial presentation, she developed diplopia, ptosis, blurred vision, slurred
speech, and difficulty walking. The patient was hospitalized with an acute delirium and multifocal neurological deficits. Most of the deficits, except the
ataxia, cleared spontaneously.
When she presented to our institution, she was alert and oriented. Visual
fields were full, and visual acuity was
20/20 bilaterally. Results of the funduscopic examination were unremarkable.
There was left beating nystagmus in the
primary gaze position; however, extraocular movements were full. All
other cranial nerves were intact. There
was a left pronator drift with 4/5
strength on her right side. Occasional
myoclonic jerks were noted. A sensory
examination revealed nothing abnormal. The patient displayed mild evidence of limb dysmetria and severe gait
ataxia. T2-weighted magnetic resonance
imaging scans revealed a hyperintense
signal in the midbrain (Fig. 1A) that extended to the diencephalon and left
temporal lobe; however, no evidence of
hemorrhage or of right temporal cortical
or subcortical lesions was observed (Fig.
1B).
Analysis of cerebrospinal fluid obtained via lumbar puncture revealed a
pressure of 14.5 cm H2O with a glucose
level of 53 mg/dl, protein level of 45
mg/dl, 38 nucleated cells/mm3 (13%
granulocytes and 78% lymphocytes),
and 88 red blood cells/mm3. Gram’s
stain and other infectious disease assays
disclosed nothing abnormal. Weak oligoclonal bands were detected. A threevessel cerebral angiogram revealed normal cerebral vasculature (Fig. 2, A and
B). A brain positron emission tomographic scan revealed hypometabolism
in the brainstem and diencephalon suggestive of an inflammatory or demyelinating lesion. The patient was treated
with methylprednisolone (1.0 g/d) for 5
days, then tapered via orally administered corticosteroids. Over the course of

1138

Bulsara et al.
FIGURE 1. Axial
T2-weighted magnetic
resonance imaging scans,
obtained at the initial
presentation of the
patient, demonstrating
hyperintense signal in the
brainstem and diencephalon (A). There is no evidence of a right posterior
temporal vascular lesion
(B).
FIGURE 2. Lateral (A)
and anteroposterior (B)
cerebral angiograms of
the right carotid artery,
obtained at the initial
presentation of the
patient, revealing no evidence of vasculitis, aneurysm, or vascular
malformation.

several weeks, the majority of her symptoms resolved, and thereafter, she remained neurologically well.
Six years later, at the age of 32, the
patient presented to the emergency
room with what she characterized as the
worst headache of her life; it was associated with nausea and vomiting. No
associated focal neurological deficit was
present.
The patient’s cranial computed tomographic and subsequent magnetic resonance imaging scans (Fig. 3) revealed a
right posterior temporal intraparenchymal hemorrhage. Cerebral angiography
revealed a new 3- by 2-cm right posterior temporal AVM fed exclusively by
the right middle cerebral artery; its primary drainage was via the vein of Labbé
(Fig. 4, A and B). There was no evidence
of deep venous drainage. The patient
was taken to the operating room, where
she underwent a right temporal craniotomy. Three-dimensional image guidance was used to localize the AVM,
which did not present directly to the
pial surface. The AVM was resected via
standard microsurgical techniques.
Postoperative angiograms revealed no
residual AVM (Fig. 5, A and B), and the
patient displayed no postoperative neurological deficits.

DISCUSSION
To our knowledge, this is the first
report of an angiographically proven de
novo intracranial AVM in an adult patient who did not have a previous neurovascular abnormality. It challenges
the widespread belief that all AVMs
have a congenital cause. Traditionally,
AVMs are thought to arise between
Weeks 4 and 8 of embryonic life (8). The
primary pathological lesion consists of
direct connections between arterial inflow and venous outflow without an
intervening capillary bed. These malformations are most common in the distribution of the middle cerebral artery (8).
Unlike other congenital vascular malformations, such as vein of Galen malformations (3, 33), no cerebral AVMs
have been detected in utero (27). This
challenges the idea that all AVMs may
arise from embryonic aberrations at the
stage of vessel formation (23). Furthermore, no evidence exists to support the
belief that AVMs arise during the late
fetal period or immediate postpartum
life (21, 27).
It is clear that not all vascular malformations have a congenital cause. Dural
AVMs are acquired after sinus thrombosis (2, 10, 13, 16, 22) or spontaneously
Neurosurgery, Vol. 50, No. 5, May 2002

FIGURE 3. Axial magnetic resonance
imaging scan of the brain, obtained
approximately 6 years after the initial
presentation, demonstrating a new right
posterior temporal intraparenchymal
hemorrhage.
(32). The senior author (MJA) previously described a de novo cerebral cavernous malformation that developed after radiation for a pituitary tumor (1).
Likewise, de novo cavernous malformations of the spinal cord have been reported after spinal axis radiation (18).
The literature involving de novo cerebral AVMs is more sparse. Schmit et al.
(26) reported an angiographically
proven acquired cerebral AVM in a
child with moyamoya disease occurring
after cerebral infarction. RodriguezArias et al. (24) reported recurrence of a
cerebral AVM in a child in a different
location after radiosurgery. More recently, local venous hypertension associated with a dural arteriovenous fistula
was reported to be associated with a de
novo cerebellar AVM (6).
An interesting associated question is
whether AVMs with angiographically
verified complete resection recur, or
whether an undiagnosed residual AVM
exists. Some reports have noted angiographic complete resection of cerebral
AVMs with subsequent recurrence, particularly in children (7, 15). This calls
into question whether a milieu of local
cytokines promotes AVM development
or growth.
Growing support argues that spinal
AVMs can arise after trauma, inflammation, infection, or various compressive
lesions (11). These processes stimulate
angiogenesis (5). Potent angiogenic factors that are abundant in ischemic brain

De Novo Cerebral Arteriovenous Malformation
FIGURE 4. Lateral (A) and
anteroposterior (B) cerebral angiograms of the
right internal carotid
artery, obtained at the time
of the second presentation,
revealing a new AVM fed
by the middle cerebral
artery with superficial
venous drainage.

1139

cally documented not to have a previous cerebrovascular abnormality. It
provides further evidence that not all
cerebral AVMs have a congenital cause.
Received, July 3, 2001.
Accepted, August 27, 2001.
Reprint requests: Michael J. Alexander,
M.D., Division of Neurosurgery, Duke University Medical Center, 4505 Busse Building,
Duke South, Durham, NC 27710.
Email: michael.alexander@duke.edu

REFERENCES

FIGURE 5. Postoperative
(A) lateral and anteroposterior (B) cerebral angiograms of the right internal
carotid artery revealing no
residual AVM.

regions are vascular endothelial growth
factor and basic fibroblast growth factor
(4, 28). It has been proposed that overstimulated angiogenesis may lead to the
development of AVMs (26, 28).
Our understanding of the genetics,
pathogenesis, and evolution of AVMs is
just beginning to develop. Germ-line
mutations with or without environmental factors affecting common angiogenic
pathways may underlie the pathophysiology of intracranial AVMs. Promising
candidate proteins range from the endothelial angiopoietin receptor Tie-2 (9,
30), transforming growth factor ␤ (12),
nitric oxide synthase (14), vascular endothelial growth factor (28, 29), and fibroblast growth factor (25).
Sure et al. (28) have demonstrated immunohistochemical staining for proliferating cell nuclear antigen, MIB-1, vascular endothelial growth factor, and its
receptor FLK-1 in the endothelium of
AVMs. This implies neoangiogenesis
and endothelial proliferation. Furthermore, increased neoangiogenesis has

been observed in partially treated
AVMs (29). The relative role of the environment in the development of these
lesions is not known, but as in the patient we describe here, environmental
factors may play a significant role in
their development.
Whatever the initial mechanism, the
increasing hemodynamic stress in the
affected vessels may cause further vascular remodeling and secondary vascular changes. This may lead to AVM maturation. Understanding the underlying
cause of AVMs may provide critical insights into potential therapeutic
management.

CONCLUSIONS
The pathogenesis of cerebral AVMs is
not fully understood, but the accepted
congenital cause of these lesions has recently been questioned. To our knowledge, this is the first demonstration of
an angiographically proven de novo cerebral AVM in a patient angiographiNeurosurgery, Vol. 50, No. 5, May 2002

1. Alexander MJ, DeSalles AA, Tomiyasu U: Multiple radiation-induced intracranial lesions after
treatment for pituitary adenoma: Case report.
J Neurosurg 88:111–115, 1998.
2. Awad IA, Little JR, Akarawi WP, Ahl J: Intracranial dural arteriovenous malformations: Factors predisposing to an aggressive neurological
course. J Neurosurg 72:839–850, 1990.
3. Campi A, Scotti G, Filippi M, Gerevini S,
Strigimi F, Lasjaunias P: Antenatal diagnosis of
vein of Galen aneurysmal malformation: MR
study of fetal brain and postnatal follow-up.
Neuroradiology 38:87–90, 1996.
4. Chen HH, Chien CH, Liu HM: Correlation between angiogenesis and basic fibroblast growth
factor expression in experimental brain infarct.
Stroke 25:1651–1657, 1994.
5. Folkman J, Kagsbrun M: Angiogenic factors.
Science 235:442–447, 1987.
6. Friedman JA, Pollock BE, Nichols DA: Development of a cerebral arteriovenous malformation in an adult documented by serial angiography. Presented at the AANS Meeting,
Toronto, April 21–26, 2001, p 71 (abstr).
7. Gabriel EM, Sampson JH, Wilkins RH: Recurrence of a cerebral arteriovenous malformation
after surgical excision: Case report. J Neurosurg
84:879–882, 1996.
8. Garretson HD: Intracranial arteriovenous malformations, in Wilkins RH, Rengachary SS
(eds): Neurosurgery. New York, McGraw-Hill,
1996, pp 2433–2442.
9. Hatva E, Jääskeläinen J, Hirvonen H, Alitalo K,
Haltia M: Tie endothelial cell-specific receptor tyrosine kinase is upregulated in the vasculature of
arteriovenous malformations. J Neuropathol Exp
Neurol 55:1124–1133, 1996.
10. Herman JM, Spetzler RF, Bederson JB, Kurbat
JM, Zabramski JM: Genesis of a dural arteriovenous malformation in a rat model.
J Neurosurg 83:539–545, 1995.
11. Hida K, Iwasaki Y, Isu T, Akino M, Abe H,
Tashiro K, Yoshida K, Miyasaka K, Takei H,
Abe S: Spinal dural AVM: Report of three cases
[in Japanese]. No Shinkei Geka 14:361–366,
1986.
12. Hirschi KK, Rohovsky SA, D’Amore PA: PDGF,
TGF-␤, and heterotypic cell-cell interactions
mediate endothelial cell-induced recruitment of
10T1/2 cells and their differentiation to a
smooth muscle fate. J Cell Biol 141:805–814,
1998.

1140

Bulsara et al.

13. Houser OW, Campbell JK, Campbell RJ, Sundt
TM Jr: Arteriovenous malformation affecting
the transverse dural venous sinus: An acquired
lesion. Mayo Clin Proc 54:651–661, 1979.
14. Joshi S, Young WL, Pile-Spellman J, FogartyMack P, Sciacca RR, Hacein-Bey L, Duong H,
Vulliemoz Y, Ostapkovich N, Jackson T: Intraarterial nitrovasodilators do not increase cerebral blood flow in angiographically normal territories of arteriovenous malformation patients.
Stroke 28:1115–1122, 1997.
15. Kader A, Goodrich JT, Sonstein WJ, Stein BM,
Carmel PW, Michelsen WJ: Recurrent cerebral
arteriovenous malformations after negative
postoperative angiograms. J Neurosurg 85:14–
18, 1996.
16. Lawton MT, Jacobowitz R, Spetzler RF: Redefined
role of angiogenesis in the pathogenesis of dural
arteriovenous malformations. J Neurosurg
87:267–274, 1997.
17. Lazar RM, Connaire K, Marshall RS, PileSpellman J, Hacein-Bey L, Solomon RA, Sisti
MB, Young WL, Mohr JP: Developmental deficits in adult patients with arteriovenous malformations. Arch Neurol 56:103–106, 1999.
18. Maraire JN, Abdulrauf SI, Berger S, Knisely J,
Awad IA: De novo development of a cavernous
malformation of the spinal cord following spinal axis radiation: Case report. J Neurosurg
90:234–238, 1999.
19. Mast H, Young WL, Koennecke HC, Sciacca RR,
Osipov A, Pile-Spellman J, Hacein-Bey L,
Duong H, Stein BM, Mohr JP: Risk of spontaneous haemorrhage after diagnosis of cerebral
arteriovenous malformation. Lancet 350:1065–
1068, 1997.
20. Mohr JP, Stein BM, Pile-Spellman J: Arteriovenous malformations, in Barnett JM, Mohr JP,
Stein BM, Yatsu FM (eds): Stroke, Pathophysiology, Diagnosis and Management. Philadelphia,
Churchill Livingstone, 1998, pp 725–750.
21. Nelson MD Jr, Gonzalez-Gomez I, Gilles FH:
Dyke Award: The search for human telencephalic ventriculofugal arteries. AJNR Am J
Neuroradiol 12:215–222, 1991.
22. Ozawa T, Miyasaka Y, Tanaka R, Kurata A,
Fujii K: Dural-pial arteriovenous malformation
after sinus thrombosis. Stroke 29:1721–1724,
1998.
23. Padget DH: The cranial venous system in man
in reference to development, adult configuration, and relation to the arteries. Am J Anat
98:307–355, 1956.
24. Rodriguez-Arias C, Martinez R, Rey G, Bravo
G: Recurrence in a different location of a cerebral arteriovenous malformation in a child after
radiosurgery. Childs Nerv Syst 16:363–365,
2000.
25. Rothbart D, Awad IA, Lee J, Kim J, Harbaugh R,
Criscuolo GR: Expression of angiogenic factors
and structural proteins in central nervous system vascular malformations. Neurosurgery 38:
915–925, 1996.
26. Schmit BP, Burrows PE, Kuban K, Goumnerova
L, Scott RM: Acquired cerebral arteriovenous
malformation in a child with moyamoya disease. J Neurosurg 84:677–680, 1996.
27. Stapf C, Mohr JP: New concepts in adult brain
arteriovenous malformations. Curr Opin
Neurol 13:63–67, 2000.

28. Sure U, Butz N, Schlegel J, Siegel AM, Wakat JP,
Mennel HD, Bien S, Bertalanffy H: Endothelial
proliferation, neoangiogenesis, and potential de
novo generation of cerebrovascular malformations. J Neurosurg 94:972–977, 2001.
29. Sure U, Butz N, Siegel AM, Mennel HD, Bien S,
Bertalanffy H: Treatment-induced neoangiogenesis
in cerebral arteriovenous malformations. Clin
Neurol Neurosurg 103:29–32, 2001.
30. Vikkula M, Boon LM, Carraway KL III, Calvert
JT, Diamonti AJ, Goumnerov B, Pasyk KA,
Marchuk DA, Warman ML, Cantley LC, Mulliken JB, Olsen BR: Vascular dysmorphogenesis
caused by an activating mutation in the receptor tyrosine kinase TIE2. Cell 87:1181–1190,
1996.
31. Wilkins RH: Natural history of intracranial vascular malformations: A review. Neurosurgery
16:421–430, 1985.
32. Yoshino O, Matsui H, Hirano N, Tsuji H: Acquired dural arteriovenous malformations of
the lumbar spine: Case report. Neurosurgery
42:1387–1389, 1998.
33. Yuval Y, Lerner A, Lipitz S, Rotstein Z, Hegesh
J, Achiron R: Prenatal diagnosis of vein of Galen
aneurysmal malformation: Report of two cases
with proposal for prognostic indices. Prenat
Diagn 17:972–977, 1997.

COMMENTS
Bulsara et al. describe an interesting
case that seems to demonstrate de novo
formation of a cerebral arteriovenous
malformation (AVM). Over the years,
numerous examples have been set forth
to demonstrate de novo formation of a
variety of different types of vascular
malformations. This is an interesting
topic because such cases potentially provide insight into the pathogenesis of
these various lesions. I found the angiograms presented as Figure 4 to be most
interesting. The anteroposterior view in
particular demonstrates the loose and
delicate nidal pattern that I typically associate with AVMs in children. It seems
that the classic adult AVM generally has
a much more dense or compact nidus.
This distinction between childhood and
adult AVMs has been made by many
clinicians over the years, and I would
currently characterize it as part of the
conventional wisdom regarding AVMs.
Significantly, the current example bears
a strong resemblance to what we might
consider a juvenile AVM, although here,
it appears in an adult. Given this observation, I am particularly disappointed
that we have so little information regarding the intraoperative appearance
of the AVM—and in particular, we are
provided no information regarding
Neurosurgery, Vol. 50, No. 5, May 2002

the histological examination of the
malformation.
I must question the assertion of
Bulsara et al. that the original angiogram is completely normal. If one looks
closely at their Figure 4A, observing the
lateral projection that demonstrates the
malformation and then mentally superimposing the malformation onto their
Figure 2A, the initial lateral projection,
one can observe a faint silhouette of the
lesion on the original study. I can observe no such faint silhouette on the
anteroposterior projections. I suspect
that this is because the malformation is
essentially tubular. The lesion is so faint
on the original films that only when
viewing it down its long axis can one
appreciate the faint blush.
If the lesion was present 6 years earlier but was simply extremely faint on
radiographic scans, what does this
mean? I suspect that all AVMs undergo
remodeling over the years. Although
numerous stimuli for such remodeling
probably exist, I would suspect that hemodynamic forces are dominant, and
among these, I would conjecture that
shear force at the level of the endothelial
wall is the most critical. It seems reasonable to think that this type of dynamic
remodeling over time is what turns the
delicate, loose juvenile AVM into a
more dense, compact AVM in adults. It
is conceivable that some early lesions
move in the opposite direction and actually regress over time. Obviously, this
remodeling does not occur by magic; it
is most likely the result of local release
of angiogenic factors. A fundamental
question is whether such a dynamic remodeling process can occur within a totally normal vascular bed, or whether
some congenital abnormality of the infrastructure must be present to promote
such remodeling. I submit that, in the
current case, we have angiographic evidence, although subtle, that an abnormality did exist before the patient’s presentation at age 32.
Robert E. Breeze
Denver, Colorado
In this fascinating case report, Bulsara
et al. describe a case of de novo formation
of a small right temporal AVM in a
woman who had had a documented ce-

De Novo Cerebral Arteriovenous Malformation
rebral angiogram that displayed nothing
abnormal 6 years earlier. This finding obviously challenges the prevalent existing
theories that pial AVMs arise from an abnormality in embryological cerebrovascular development, and it raises the possibility that at least some pial AVMs arise
later in life, perhaps as a consequence of
environmental or genetic factors in a
manner analogous to dural AVMs or cavernous hemangiomas.
Robert Replogle
Duke S. Samson
Dallas, Texas
Bulsara et al. provide excellent detail
and analysis in a novel observation. This
patient clearly experienced an episode of
meningoencephalitis, which fortunately
was investigated angiographically. Six
years later, she bled from a de novo AVM.
If one studies the original arteriogram
carefully, on the lateral projection at the
precise site that the new AVM developed,
one sees a slight increase in vascularity.
There is certainly no shunting, but in retrospect, there is possibly an apparent angiographic abnormality. The authors’ discussion and review of the literature are
insightful and provocative. I am aware of
two children in my own practice whose

postoperative angiograms revealed nothing abnormal but who bled several years
later from a new AVM. In both of these
cases, the new AVM was in the general
area of the original lesion. Nevertheless,
the potential for focal hyperangiogenesis
driven by any number of potential mechanisms could conceivably create AVMs in
adults. I suspect that this case report will
prompt all neurosurgeons to follow their
posttreatment AVM patients a little more
carefully.
H. Hunt Batjer
Chicago, Illinois
This seminal case report will likely be
cited for years to come. Although de
novo genesis of cavernous malformations and dural AVMs has been well
documented, there has been no equally
convincing documentation in a cerebral
AVM. Yet the progression of these lesions and occasional spontaneous resolution have been reported, suggesting
dynamic angiogenesis within these lesions. Recurrence of AVMs after documented and verified excision has also
been reported. With this careful case report by Bulsara et al., we now have convincing evidence of de novo genesis of a
typical cerebral AVM in an adult patient

who did not harbor such a lesion a few
years earlier.
The appearance of the cerebral AVM
in this patient is characteristic of a diffuse, lacy nidus, reminiscent of AVMs in
children. Perhaps the AVM nidus becomes more compact as it matures and
as progressive ischemia and gliosis replace the brain tissue intervening between the vascular channels. Because
this lesion was not embolized preoperatively, it would be helpful to examine it
histologically in some detail, in comparison to the more conventional AVM abnormality. Also, the resolution of the
various imaging modalities does not
preclude the possibility of a preexisting
occult vascular anomaly, which then
evolved into full-blown AVM. The likelihood of activated angiogenesis in association with the patient’s previous illness leads to other speculations
regarding contributing factors that may
have triggered this unusual event. As a
result of this case report and other carefully documented observations, new hypotheses will be formulated regarding
the pathogenesis of cerebral AVMs.
Issam A. Awad
Denver, Colorado

Special Note on Color Figures
To all authors who submit manuscripts to Neurosurgery that include color figures: Please
be aware of the following policy of Neurosurgery, which is also printed in the
Information for Contributors:
Figures that are submitted in color are automatically published in color, and
authors will be responsible for the costs (typically $1200-2000 per article) of
any color reproductions. Authors who subsequently want their color figures
published in black and white will be billed for the color processing costs
(typically $300-500 per article). Authors are responsible for costs incurred to
retouch or correct submitted illustrations.
Special Note on Pathology Figures—With all histopathology, and in
particular immunohistochemistry illustrations, color is preferred.
Thank you for your cooperation in helping us maintain the high scientific quality of
Neurosurgery.

Neurosurgery, Vol. 50, No. 5, May 2002

1141