Radiosurgery for unruptured
cerebral arteriovenous malformations
Long-term seizure outcome
Seung-Yeob Yang, MD,
Objective: To date, seizures in relation to arteriovenous malformations (AVM) have been a secondary
Dong Gyu Kim, MD,
target of most studies. The insufficient evaluation, in conjunction with the lack of consistent seizure
outcome assessment, has made it been difficult to draw conclusions about seizure outcome after
Hyun-Tai Chung, PhD
radiosurgery for AVM. This study aimed to determine the effect of radiosurgery on seizure outcome
Sun Ha Paek, MD, PhD
depending on AVM obliteration and on the development of new seizure in patients with AVM.
Methods: Between 1997 and 2006, 161 consecutive patients underwent radiosurgery for un-
Correspondence & reprint
ruptured AVM and were retrospectively assessed with a mean follow-up of 89.8 months by their
requests to Dr. Kim:
medical records, updated clinical information, and, when necessary, direct patient contact. Sei-
zure outcome was assessed using the Engel seizure frequency scoring system.
Results: Of the 86 patients with a history of seizure before radiosurgery, 76.7% (66/86) were
seizure-free and 58.1% (50/86) were medication-free at the last follow-up visit. Of the patients
who achieved AVM obliteration, 96.7% (58/60) were seizure-free while 30.8% (8/26) of those
patients who did not achieve AVM obliteration were seizure-free (p
0.001). The proportion of
patients who were medication-free was 81.7% (49/60) of the patients with obliteration and
3.8% (1/26) of patients without obliteration (p
0.001). Of the 75 patients with no history of
seizure before radiosurgery, 10 had provoked seizures due to the direct and indirect radiosurgical
influences after radiosurgery.
Conclusions: Although radiosurgery tends to cause seizures temporarily, the radiosurgery may
improve seizure outcomes in patients with AVM-related seizures, especially in patients with AVM
obliteration. Neurology(R) 2012;78:1292-1298
antiepileptic drug; ARUBA
A Randomized Multicenter Clinical Trial of Unruptured Brain AVMs; AVM
venous malformation; CI
confidence interval; GK
gamma knife; OR
odds ratio; PRI
Until now, the prognosis of epilepsy-associated cerebral arteriovenous malformations (AVMs)
has received little attention. Similar to the decreased risk of hemorrhage from cerebral AVMs
following stereotactic radiosurgery,1 seizure improvement may also be common in patients
with cerebral AVMs who undergo radiosurgery.2,3 Unfortunately, most of the available data in
this field come from studies that have assessed the patients at less than 5 years following
radiosurgery, and no study has investigated the long-term seizure outcome beyond 5 years.
Moreover, it has been unclear whether, and to what extent, seizure frequency decreases both
during a latency period and after AVM obliteration, compared to seizure frequency before
radiosurgery. The impact of angiographic obliteration on seizure freedom and cessation of
Editorial, page 1286
antiepileptic drugs (AEDs) after radiosurgery also remains unknown. In order to address the
long-term seizure outcome of AVM radiosurgery and the development of new seizures after
Supplemental data at
radiosurgery, we performed a retrospective study on 161 patients who had undergone radiosur-
gery for their unruptured AVM.
From the Department of Neurosurgery (S.-Y.Y.), Dongguk University Graduate School, Seoul; and Department of Neurosurgery (D.G.K., H.-T.C.,
S.H.P.), Seoul National University College of Medicine, Seoul, Korea.
Go to Neurology.org for full disclosures. Disclosures deemed relevant by the authors, if any, are provided at the end of this article.
Copyright (c) 2012 by AAN Enterprises, Inc.
Distribution of patients by seizure frequency scoring system
For the first
At the last
Seizure-free, off antiepileptic drug
Seizure-free, need for antiepileptic drug
Seizure-free, requires antiepileptic drugs to
Nondisabling simple partial seizures
Nondisabling nocturnal seizures only
1-3 per year
4-11 per year
1-3 per month
1-6 per week
1-3 per day
4-10 per day
>10 per day but not status epilepticus
Status epilepticus without barbiturate coma
METHODS Patients. Between 1997 and 2006, 411 consec-
partial, and partial with secondary generalization or generalized
utive patients with an angiographically proven cerebral AVM
seizures. Preradiosurgical seizure frequency score was defined as
underwent gamma knife (GK) radiosurgery at our hospital. Of
the score for the year preceding radiosurgery. In order to assess
the 411 patients, 14 were lost during follow-up, and were ex-
the early effect of radiosurgery on seizure outcomes before AVM
cluded from this study because the follow-up after radiosurgery
obliteration, we investigated seizure frequency score for the first
was less than 1 year or data were insufficient to record an Engel
year after radiosurgery. For this evaluation, the "at the last
seizure frequency score (table e-1 on the Neurology(R) Web site at
follow-up" postradiosurgical outcome was assessed during the
www.neurology.org). It is difficult to precisely evaluate the effect
last 12 months preceding the analysis of data. Seizure freedom
of radiosurgery on seizure outcome in patients who had experi-
and cessation of medication were defined as seizure frequency
enced hemorrhages because of the epileptogenic effect of hemor-
scores of 0 -2 and score 0, respectively.9 To assess the effect of the
rhage.4-6 Therefore, we excluded the 236 patients who had
preradiosurgical seizure frequency or duration on seizure out-
experienced intracranial hemorrhages before radiosurgery. In all,
come, the patients who had presented with seizure were divided
161 patients were enrolled in this study to assess the long-term sei-
into 2 groups: patients with only a single seizure and patients
zure outcome and the epileptogenic effect of radiosurgery (table
with more than 1 seizure before radiosurgery.
e-1). None of these 161 patients had any other brain lesions causing
seizures, progressive CNS disorders, or psychotic problems.
Statistical analysis. The median time to achieve seizure free-
dom or cessation of medication was determined on the basis of
Standard protocol approvals, registrations, and patient
Kaplan-Meier event-free survival curves. Associations between
consents. The Institutional Review Board of our hospital ap-
categorical variables and seizure outcome were analyzed using
proved all aspects of the study. However, it did not require in-
Pearson 2 test (or Fisher exact test as appropriate). Continuous
formed consent from these patients for their inclusion because
variables were compared using Student t tests. Prognostic factors
the study depended only on information obtained as a part of
considered for univariate analysis included patient demograph-
routine clinical care and patient medical records.
ics, Spetzler-Martin grade, seizure frequency score before radio-
Radiosurgical technique and follow-up evaluations.
surgery, single seizure only before radiosurgery, embolization
The radiosurgical technique used was reported previously.7 De-
before radiosurgery, duration of epilepsy before radiosurgery,
tailed description of the follow-up evaluations is provided as ap-
type and onset of seizures, switch or increment of AEDs after
pendix e-1. AVM characteristics and radiosurgical parameters
radiosurgery, and AVM obliteration after radiosurgery. Multi-
are summarized in table e-1.
variable logistic regression analysis was used to assess the associa-
The postradiosurgery imaging (PRI) changes were defined as
tion of variables with seizure freedom or off medication. To
the development of signal changes on follow-up T2-weighted or
control multicollinearity, radiation dose, AVM volume, and
fluid-attenuated inversion recovery MRI, regardless of whether
AVM size from the regression analysis (all variance inflation fac-
these changes were accompanied by symptoms. The severity of
10.0) were omitted. Results are presented as ad-
PRI changes was classified as mild, moderate, or severe according
justed odds ratios (OR) with a 95% confidence interval (CI).
to our previous report.8
The level of statistical significance was set as p
Assessment of seizure outcomes. We assessed seizure out-
come by comparing the preradiosurgical and postradiosurgical
RESULTS Seizure development after radiosurgery.
seizure frequencies using the Engel seizure frequency scoring sys-
Of the 75 patients with no history of seizures before
tem (table 1).9 Seizures were classified as simple partial, complex
radiosurgery, no patient was on AEDs before and at
Neurology 78 April 24, 2012
Seizure outcome after radiosurgery (n
(A) The probability of seizure freedom (54.7% at 2 years and 73.3% at 5 years). (B) The probability of off medication
(30.2% at 2 years and 54.7% at 5 years).
the time of radiosurgery. The mean clinical and neu-
The distribution of patients by seizure frequency
roradiologic follow-up periods after radiosurgery
score before radiosurgery, for the first year, and at the
were 89.2 and 56.2 months, respectively. At the last
last follow-up are presented in table 1. For the first
follow-up, angiographically proven AVM oblitera-
year after radiosurgery, 27.9% (24/86) of patients
tion was observed in 61 of the 75 patients. Of the 75
were seizure-free while 3.5% (3/86) of patients had
patients, 10 had provoked seizures after radiosurgery.
an increased seizure frequency score. However, there
Of these 10 patients, 1 developed a partial seizure
was a difference in the proportion with seizure free-
within 24 hours after radiosurgery without neuroim-
dom ( p
0.001). Of the 3 patients with increased
aging changes, but no recurrence was seen without
seizure frequency score for the first year, only 1 re-
medication ever since; 5 had provoked seizures sev-
ported a worsening of seizure at the last follow-up.
eral months to years after radiosurgery due to severe
The seizure frequency scores were different between
PRI changes, but no recurrence was seen after the
before radiosurgery and at last follow-up ( p
resolution of the PRI changes; and 4 had provoked
0.001). At the last follow-up, 76.7% (66/86) of pa-
seizures due to latency period hemorrhages.
tients were seizure-free; 96.7% (58/60) of patients
Seizure freedom. Of the 86 patients with a history of
who achieved AVM obliteration were seizure-free;
seizures before radiosurgery, 72 (84%) were followed
and 30.8% (8/26) of those who did not achieve
for at least 5 years. The mean clinical and neurora-
AVM obliteration were seizure-free ( p
diologic follow-up periods after radiosurgery were
tients with AVM obliteration remained seizure-free
90.2 and 57.5 months, respectively. At the last
at the last follow-up, but 2 of the 26 patients who did
follow-up, angiographically proven AVM oblitera-
not experience AVM obliteration had a relapse of
tion was observed in 60 of the 86 patients.
seizures during AED tapering. The median time to
Provoked seizures occurred in 3 patients within
achieve seizure freedom after radiosurgery was 2.0
24 hours after radiosurgery without apparent neuro-
years (95% CI 1.5-2.5) (figure, A).
imaging changes, but all of these patients remained
The results after univariate analysis based on dif-
seizure-free subsequently. Four patients had in-
ferent parameters are given in table 2. Multivariate
creased seizure frequency several months to years af-
analysis defined 2 independent prognostic factors
ter radiosurgery probably due to severe PRI changes,
predicting seizure freedom: AVM obliteration ( p
but no recurrence was seen after the resolution of the
0.001) and seizure frequency score before radiosur-
PRI changes. At the last follow-up, the 4 patients
gery ( p
0.011) (table 2).
were medication-free, and they all achieved AVM
Prior to radiosurgery, 36% (31/86) of patients
obliteration. The PRI changes were not associated
had a single seizure only; the remainder of the pa-
with either seizure freedom ( p
0.376) or going off
tients had more than one seizure. At the last follow-
medication ( p
0.309) at the last follow-up.
up, 90.3% (28/31) of the patients with only a single
Neurology 78 April 24, 2012
seizure were seizure-free; 100% (25/25) of patients
who achieved AVM obliteration were seizure-free;
and 50% (3/6) of those who did not achieve AVM
obliteration were seizure-free ( p
0.004) (table 3).
Patients with only a single seizure tended to achieve a
higher seizure-free rate than patients with more than
one seizure (90.3% vs 69.1%); however, this differ-
ence was not significant in the multivariate analysis
0.055). Although the correlation between
seizure-free and seizure duration before radiosurgery
was not significant, there was a tendency for a shorter
mean duration in seizure-free patients than in those
who were not seizure-free at the last follow-up (35.3
months vs 49.6 months, respectively).
Off medication. Of the 86 patients with a history of
seizures before radiosurgery, all patients were on
AEDs before radiosurgery: 62 were taking one kind
of AED, 21 were taking 2 kinds of AEDs, and 3 were
taking 3 kinds of AEDs. At the last follow-up, 50
patients were off AEDs, 23 were taking one kind of
AED, 11 were taking 2 kinds of AEDs, and 2 were
taking 3 kinds of AEDs. However, either AEDs were
switched or their doses were increased in 6 patients:
AEDs were switched once in 4 patients, twice in one
patient, and doses were increased in another patient.
AED modification was not associated with seizure
freedom ( p
0.353). At the last follow-up, 50 of the
patients were medication-free, whose follow-up pe-
riod after cessation of medication was from 4 to 133
months (mean 67 months). The proportions of
medication-free patients with and without an AVM
obliteration were 81.7% (49/60) and 3.8% (1/26),
respectively ( p
0.001). The median time to
achieve the cessation of medication after radiosurgery
was 4.0 years (95% CI 2.8 -5.2) (figure, B).
The results after univariate analysis based on
different parameters are given in table 3. Multivar-
iate analysis revealed that AVM obliteration ( p
0.001) and seizure frequency score before radio-
surgery ( p
0.045) were associated with going off
medication (table 2).
At the last follow-up, 74.2% (23/31) of the pa-
tients with only a single seizure were off AEDs; 92%
(23/25) of patients who achieved AVM obliteration
were off AEDs; and 0% (0/6) of those who did not
achieve AVM obliteration were off AEDs ( p
0.001) (table 3).
radiosurgery-related complications is provided as
DISCUSSION We found that radiosurgery is effec-
tive for seizure control even before AVM oblitera-
tion, although the radiosurgery tends to cause
Neurology 78 April 24, 2012
Relation between single seizure only before RS and seizure outcomea
Patients with only a single seizure before
Patients with more than one seizure before
RS (N 55)
( ) (n
( ) (n
( ) (n
( ) (n
25 (100)/0 (0)
3 (50.0)/3 (50.0)
33 (94.3)/2 (5.7)
5 (25.0)/15 (75.0)
23 (92.0)/2 (8.0)
0 (0)/6 (100)
26 (74.3)/9 (25.7)
1 (5.0)/19 (95.0)
arteriovenous malformation; RS
a Data are number (%) of patients.
seizures temporarily. A number of studies have dem-
correlation between seizure outcome and AVM
onstrated that the seizure frequency begins to de-
obliteration, but that there is a tendency for a better
crease several months after radiosurgery and that the
seizure outcome for patients with obliterated AVMs.
reduction of seizure frequency and intensity does not
A possible explanation may simply be that while they
seem to require AVM obliteration or morphologic
analyzed patients at the fixed time points of 1 and 3
changes.2,6,10-13 The results from our study are con-
years following radiosurgery, our study was focused
sistent with these results. Although all patients did
on the patients with more than 4 years of follow-up
not experience AVM obliteration 1 year after radio-
after radiosurgery. They showed that AVM diameter
surgery, 24 patients became seizure-free, and 19 of
is associated with an excellent seizure outcome not at
these 24 patients achieved cessation of medication.
the first year of follow-up, but at year 3 of follow-up.
At the last follow-up, 8 of 26 patients without AVM
They also noted that longer follow-up is necessary to
obliteration became seizure-free, and 1 of these 8 pa-
more accurate assessment of the role of radiosurgery.
tients discontinued medication. These findings sug-
Prior studies have shown that patients with longer
gest that radiosurgery can reduce seizure frequency
seizure history and frequent seizures had a poorer sei-
and intensity via some intrinsic effects on the AVM
zure outcome than those without longer seizure his-
nidus and surrounding area.6,12,13 The suppression of
tory and frequent seizures.2,3,19-21 Our study is
epileptic activity by a neuromodulatory effect at non-
consistent with these results, which seems to be asso-
necrotizing radiation doses has been proposed as pos-
ciated with the development of secondary epilepto-
sible intrinsic effects.13-16 Biochemically, radiation
genic foci distant from the AVM that, consequently,
has been hypothesized to inhibit protein synthesis,
were not irradiated. Secondary epileptogenesis has
preventing the maintenance of spontaneous neuron
been already described in patients with epilepsy
bursts,16 and have differential effects on the inhibi-
caused by AVMs.19-21
tory GABA system and the excitatory amino acid sys-
Seizure improvement is quite common in patients
tem.14,15 Accordingly, it might be possible to modify
with AVMs who have undergone radiosurgery and
the brain region adjacent to the AVM to make it
rivals the results provided by microsurgical resection
nonepileptic while preserving its functional role.14,15
of AVMs2,3,6,11,13,17,18,21,22; 70.4%- 83.0% of AVM
In our study, 27.9% of patients were seizure-free 1
patients with preoperative seizure became seizure-
year after radiosurgery, and 76.7% of patients were
free or had occasional auras after microsurgical re-
seizure-free at the last follow-up. These findings sug-
section.21,22 The seizure improvement achieved by
gest ongoing changes after radiosurgery in the AVM
radiosurgery warrants the use of this technique in
nidus and surrounding area. Recent studies have re-
patients with medically refractory seizures, but
ported that more patients with AVM obliteration
microsurgery is perhaps preferable to radiosurgery
were seizure-free than patients without.3,17 The re-
in terms of the immediate prevention of further hem-
sults from our study show a similar outcome. Based
orrhages.21,22 However, in addition to microsurgery-
on our findings and those of others,3,17,18 although
related complications, microsurgical resection may
radiosurgery has beneficial effects on seizure out-
create a new seizure focus and may increase seizure
comes even before AVM obliteration, higher seizure-
frequency due to cortical or subcortical damage by
free rates were observed in patients with AVM
manipulation; 16.3-31.6% of AVM patients with-
obliteration. In addition to the neuromodulatory ef-
out a history of preoperative seizures had new onset
fect of irradiation mentioned above, radiosurgical re-
seizures.22,23 In our study, mean radiologic follow-up
duction of the steal phenomenon may additionally
was 56.8 months during which 10 instances of AVM
contribute to the reduction of epileptogenic activity
rupture occurred leading to a hemorrhage rate of
in the ischemic areas surrounding an AVM.3,10,11,17,18
minimum 1.3% per year until obliteration; compara-
However, a recent study2 showed that there is no
ble findings have also been reported.2,17 This suggests
Neurology 78 April 24, 2012
slightly higher bleeding risk in unruptured brain
The gold standard for evaluating the effect of radio-
AVMs following radiosurgery as compared to natural
surgery on seizure outcome would be a randomized trial
history.24-26 These series on the natural history reflect
comparing a group undergoing radiosurgery with a
a selection bias regarding AVM size, location, and
group receiving medication only. However, there have
other clinical and anatomic characteristics and com-
not been any randomized trials comparing any of the
parisons between published data and our series have
forms of interventional therapy (with endovascular pro-
to be made with caution.
cedures, microsurgery, or radiosurgery, alone or in
A recent study reported that 13 of the 65 patients
combination) for AVMs among themselves or with
examined had provoked seizures several months to
medical management. A new trial, A Randomized
years after radiosurgery due to low AED levels or due
Multicenter Clinical Trial of Unruptured Brain
to edema as confirmed by neuroimaging.2 In another
AVMs (ARUBA),27 began in 2006 to determine
study, 10 of the 80 patients with PRI changes had
whether medical management improves long-term
provoked seizures; however, seizures were controlled
outcomes of patients with unruptured AVMs com-
with or without AEDs, and no other intervention
pared to interventional therapy, and is ongoing. The
was needed.5 In our study, 9 of the 161 patients had
ARUBA dataset includes data on seizure recur-
problems with new or increased seizure activity that
rence on prospective follow-up in patients with or
related to PRI changes, but the PRI changes were not
without complete AVM obliteration. According to
ultimately associated with seizure outcomes at the
the supporting data of the ARUBA study, radio-
last follow-up. Based on the findings of our study
surgery may increase the risk of hemorrhage and
and those of others,2,5 although the PRI changes may
induce a disabling persisting clinical syndrome in
be an important seizure-inducing factor, this effect
AVM patients who have not yet bled. Neverthe-
seems to be working temporarily until the resolution
less, while we await the results of the ongoing
of the PRI changes.
ARUBA study, we consider radiosurgery in
Our study has some limitations that need to be
patients with a cerebral AVM because of the
considered. Due to treatment selection (AVMs treat-
beneficial effects of radiosurgery in terms of angio-
able by radiosurgery) and referral pattern, clinical pa-
graphic cure, prevention of hemorrhage, and im-
tient characteristics may be subject to systematic bias
provement of seizure outcome.1-3,6,11,13,17 We
and influence the outcome of the analysis.
performed a retrospective observational study and
To analyze the association between AVM radio-
did not include a control group of patients; neverthe-
surgery and seizures, a direct comparison of 2 groups'
less, the relatively large size and long-term
outcomes (ruptured vs unruptured AVM before ra-
follow-up of our cohort make it well-suited to an
diosurgery) could provide more information. How-
assessment of seizure outcomes of radiosurgery.
ever, seizure focus can be generated there by
Also, the results of this study will be interesting to
mechanical compression, ischemic insult, and stimu-
compare to the results of the ARUBA study.
lation by hematoma or hemosiderin deposits and by
resultant gliosis in surrounding brain.4-6 Because of
the epileptogenic effect of hemorrhage as above, it is
Dr. Yang and Dr. Kim developed the hypothesis for this study and wrote
hard to assess the clear association between the AVM
the first draft. Analysis of clinical data was done principally by Dr. Yang,
Dr. Kim, and Dr. Chung. All authors contributed to data interpretation.
radiosurgery and seizures. Therefore, the direct com-
Dr. Yang, Dr. Kim, Dr. Paek, and Dr. Chung critically revised the first
parison between 2 groups will be left an area for fu-
draft and approved the final manuscript.
A potential problem is the delay in confirming
AVM obliteration. The exact date of AVM oblitera-
The authors thank Dr. Sang Gun Lee (Seoul National University, Seoul)
tion was unclear because obliteration was only iden-
for providing insight in the planning stages of the study; Dr. Jung Ho
Han and Sang Sun Chung, RN (Seoul National University Hospital,
tified at the time of angiography. Serial imaging was
Seoul), who collected data for this study; and Dr. Byung Joo Park (Seoul
performed every 6 months, and angiography was rec-
National University, Seoul) and the Medical Research Collaborating
ommended at 3 years after radiosurgery. For patients
Center at Seoul National University Hospital for their assistance in
whose AVM had disappeared on CT or MRI after
statistical analysis. Dr. Moon Hee Han (Seoul National University,
Seoul) helped us review the neuroradiologic findings of cerebral arte-
radiosurgery, angiography was performed earlier.
Therefore, we assume that AVM obliteration oc-
curred a maximum of 6 months earlier than its con-
firmation by angiography. For this reason, we
The authors report no disclosures relevant to the manuscript. Go to
analyzed the data, including the data during the first
Neurology.org for full disclosures.
year after radiosurgery, to assess the effect of radio-
surgery on seizure outcome as the only factor.
Received April 28, 2011. Accepted in final form August 12, 2011.
Neurology 78 April 24, 2012
Regis J, Kerkerian-Legoff L, Rey M, et al. First biochemi-
Maruyama K, Kawahara N, Shin M, et al. The risk of
cal evidence of differential functional effects following
hemorrhage after radiosurgery for cerebral arteriovenous
gamma knife surgery. Stereotact Funct Neurosurg 1996;
malformations. New Engl J Med 2005;352:146 -153.
66(suppl 1):29 -38.
Schauble B, Cascino GD, Pollock BE, et al. Seizure out-
Regis J, Bartolomei F, Hayashi M, Chauvel P. Gamma
comes after stereotactic radiosurgery for cerebral arterio-
knife surgery, a neuromodulation therapy in epilepsy sur-
venous malformations. Neurology 2004;63:683- 687.
gery! Acta Neurochir Suppl 2002;84:37- 47.
Kurita H, Kawamoto S, Suzuki I, et al. Control of epilepsy
Chalifoux R, Elisevich K. Effect of ionizing radiation on
associated with cerebral arteriovenous malformations after
partial seizures attributable to malignant cerebral tumors.
radiosurgery. J Neurol Neurosurg Psychiatry 1998;65:
Stereotact Funct Neurosurg 1996;67:169 -182.
648 - 655.
Lim YJ, Lee CY, Koh JS, Kim TS, Kim GK, Rhee BA.
Wolf HK, Roos D, Blumcke I, Pietsch T, Wiestler OD.
Seizure control of Gamma Knife radiosurgery for non-
Perilesional neurochemical changes in focal epilepsies.
hemorrhagic arteriovenous malformations. Acta Neuro-
Acta Neuropathol 1996;91:376 -384.
chir Suppl 2006;99:97-101.
Flickinger JC, Kondziolka D, Lunsford LD, et al. A multi-
Hoh BL, Chapman PH, Loeffler JS, Carter BS, Ogilvy CS.
institutional analysis of complication outcomes after arte-
Results of multimodality treatment for 141 patients with
riovenous malformation radiosurgery. Int J Radiat Oncol
brain arteriovenous malformations and seizures: factors as-
Biol Phys 1999;44:67-74.
sociated with seizure incidence and seizure outcomes. Neu-
Kida Y, Kobayashi T, Tanaka T, Mori Y, Hasegawa T,
rosurgery 2002;51:303-309; discussion 309 -311.
Kondoh T. Seizure control after radiosurgery on cerebral
Murphy MJ. Long-term follow-up of seizures associated
arteriovenous malformations. J Clin Neurosci 2000;
with cerebral arteriovenous malformations: results of ther-
7(suppl 1):6 -9.
apy. Arch Neurol 1985;42:477- 479.
Yang SY, Kim DG, Chung HT, Paek SH, Park JH, Han
Yeh HS, Privitera MD. Secondary epileptogenesis in cere-
DH. Radiosurgery for large cerebral arteriovenous malfor-
bral arteriovenous malformations. Arch Neurol 1991;48:
mations. Acta Neurochir 2009;151:113-124.
Han JH, Kim DG, Chung HT, et al. Clinical and neuro-
Yeh HS, Tew JM Jr, Gartner M. Seizure control after sur-
imaging outcome of cerebral arteriovenous malformations
gery on cerebral arteriovenous malformations. J Neurosurg
after gamma knife surgery: analysis of the radiation injury
rate depending on the arteriovenous malformation vol-
Piepgras DG, Sundt TM Jr, Ragoowansi AT, Stevens L.
ume. J Neurosurg 2008;109:191-198.
Seizure outcome in patients with surgically treated cerebral
Engel J Jr, Van Ness PC, Rasmussen TB, Ojemann LM.
Outcome with respect to epileptic seizures. In: Engel J Jr,
arteriovenous malformations. J Neurosurg 1993;78:5-11.
ed. Surgical Treatment of the Epilepsies, 2nd ed. New
Thorpe ML, Cordato DJ, Morgan MK, Herkes GK. Post-
York: Raven Press; 1993: 609 - 621.
operative seizure outcome in a series of 114 patients with
Heikkinen ER, Konnov B, Melnikov L, et al. Relief of
supratentorial arteriovenous malformations. J Clin Neuro-
epilepsy by radiosurgery of cerebral arteriovenous malfor-
mations. Stereotact Funct Neurosurg 1989;53:157-166.
Stapf C, Mast H, Sciacca RR, et al. Predictors of hemor-
Gerszten PC, Adelson PD, Kondziolka D, Flickinger JC,
rhage in patients with untreated brain arteriovenous mal-
Lunsford LD. Seizure outcome in children treated for arte-
formation. Neurology 2006;66:1350 -1355.
riovenous malformations using gamma knife radiosurgery.
Stapf C. [Unruptured brain arteriovenous malformations.]
Pediatr Neurosurg 1996;24:139 -144.
Rev Neurol 2008;164:787-792.
Steiner L, Lindquist C, Adler JR, Torner JC, Alves W,
Wedderburn CJ, van Beijnum J, Bhattacharya JJ, et al.
Steiner M. Clinical outcome of radiosurgery for cerebral
Outcome after interventional or conservative management
arteriovenous malformations. J Neurosurg 1992;77:1- 8.
of unruptured brain arteriovenous malformations: a pro-
Eisenschenk S, Gilmore RL, Friedman WA, Henchey RA.
spective, population-based cohort study. Lancet Neurol
The effect of LINAC stereotactic radiosurgery on epilepsy
associated with arteriovenous malformations. Stereotact
The ARUBA Trial. Available at: http://www.arubastudy.
Funct Neurosurg 1998;71:51- 61.
org/index.html. Accessed April 10, 2011.
Neurology(R) Launches Subspecialty Alerts by E-mail!
Customize your online journal experience by signing up for e-mail alerts related to your subspecialty
or area of interest. Access this free service by visiting http://www.neurology.org/site/subscriptions/
etoc.xhtml or click on the "E-mail Alerts" link on the home page. An extensive list of subspecialties,
methods, and study design choices will be available for you to choose from--allowing you priority
alerts to cutting-edge research in your field!
Neurology 78 April 24, 2012