J Neurosurg 78:12-18, 1993

Seizure control after surgery on cerebral arteriovenous
malformations

Hwa-sHaIn YEH, M.D., Joun M. Tew, JR., M.D., AND
MAUREEN GARTNER, B.S.N., M.Ep.

Department of Neurosurgery, University of Cincinnati College of Medicine and Mayfield Neurological
Institute, Cincinnati, Ohio

Prediction of seizure control after surgery on cerebral arteriovenous malformations (AVM's) is currently
unavailable. Between 1982 and 1990, 54 patients (30 males, 24 females) with epilepsy caused by a supratentorial
cerebral AVM, without prior manifestation of intracranial hemorrhage, were surgically treated. Patients ranged
in age from 11 to 59 years at seizure onset and from 13 to 70 years at surgery; the duration of seizure history
ranged from several months to 27 years. The AVM’s were located in the temporal (17 cases), frontal (15),
parietal (10), rolandic (two), and occipital (two) regions; eight were multilobular. All patients underwent
preoperative electroencephalography, intraoperative electrocorticography, and total excision of the AVM;
additional cortical excision was performed in 25 cases. Remote seizure foci were identified in the ipsilateral
mesial temporal structure in 10 patients with AVM’s located in the lateral or posterior temporal lobe and in
one with an AVM in the anterior frontal region. Two patients required a second operation to remove a remote
seizure focus. Among the 54 patients, there were no operative deaths. After surgical treatment, two patients
developed hemiparesis, one had contralateral paresthesia of limbs, two suffered partial visual field defects, and
five experienced temporary speech disturbances.

Postoperative results of seizure control during follow-up study (mean duration 4.8 years) were excellent in
38 patients (70.4%), good in 10 (18.5%), fair in five (9.3%), and poor in one (1.9%). Results appear to correlate
with age at seizure onset, duration of seizures, location of lesions, and cortical excision. Excellent results were
shown in 18 (60%) of 30 patients whose age at seizure onset was 30 years or less and in 20 (83.3%) of 24
whose age at seizure onset was greater than 30 years. Eighteen (90%) of 20 patients had excellent results when
seizure duration was | year or less; only 25% of these underwent cortical excision. Twelve (71%) of the 17
temporal AVM’s were associated with demonstrable epileptic foci. Secondary epileptogenesis can occur in
humans with supratentorial cerebral AVM’s; cortical excision in selected patients can improve the autcome of
seizure control. Early surgery of a cerebral AVM in young patients presenting with epilepsy is an important
consideration.
seizure «* secondary epileptegenesis +

Key Worps ~* arteriovenous malformation «+

outcome

E report the results of surgical treatment in
W 27 patients with cerebral arteriovenous mal-
formations (AVM’s) who presented with epi-
lepsy and no clinical history of intracranial hemor-
rhage.”° We review the follow-up data of these patients
plus 27 additional patients to assess the outcome after
surgical excision of AVM’s and to discuss the value of
further resection of epileptic foci.

Clinical Material and Methods
Patient Population

In this study, 54 patients underwent surgical excision
of supratentorial cerebral AVM’s with or without ad-
ditional resection of epileptic foci. Procedures were

12

performed at the University of Cincinnati Hospital
between January, 1982, and March, 1990. All patients
presented with epilepsy and with no clinical history of
intracranial hemorrhage. Patients were evaluated ac-
cording to the same protocol of preoperative assessment
and intraoperative management described previously”
and were followed periodically.

The patients, 30 males and 24 females, ranged in age
at surgery from 13 to 70 years; 17 patients were aged
31 to 40 years (Table 1). Patients were aged 11 to 59
years at seizure onset; the duration of seizure history
varied from a few months to 27 years. Eighteen patients
had simple partial seizures, 34 had complex partial
seizures, six had generalized seizures without aura or
focal symptom, and 22 had partial seizures with sec-

J. Neurosurg. / Volume 78 / January, 1993
Seizure control after AVM surgery

TABLE |
Age at surgery in 54 patients with cerebral arteriovenous
malformations
Age Range No. of _ Sex
(yrs) Cases Male Female
11-20 7 4 3
21-30 9 6 3
31-40 7 8 9
41-50 14 9 5
51-70 7 3 4
total cases 54 30 24
TABLE 2

Seizure type correlated with location of AVM in 54 patients*

No, of Partial Seizure Generalized Seizure

Location
of AVM — Cases

Simple Complex Primary Secondary

frontal 15 2 8 5 4
temporal 7 1 16 0 9
rolandic 2 0 2 0 1
parietal 10 8 2 0 4
occipital 2 2 0 id 1
multitobular 8 5 6 1 3
totals 54 18 34 6 22

* Abbreviation: AVM = arteriovenous malformation.

ondary generalization (Table 2). Twenty-two patients
suffered two types of seizures and two patients had
three types of seizures.

Diagnosis

Cerebral angiography, computerized tomography,
and/or magnetic resonance imaging of the brain were
performed for all patients. All 54 cerebral AVM’s were
supratentorial, with 27 involving the left and 27 the
right cerebral hemispheres. In location, 17 (31.4%) of
the 54 AVM’s were temporal, 15 (27.8%) frontal, 10
(18.6%) parietal, eight (14.8%) multilobular, two (3.7%)
rolandic, and two (3.7%) occipital. The AVM was
characterized as large if it measured greater than 4 cm
in diameter (23 cases), medium-sized if between 2 and
4 cm in diameter (19), and small if less than 2 cm in
diameter (12).

Electroencephalographic Study

All patients underwent preoperative electroencepha-
lographic (EEG) recording. Sphenoidal electrodes were
inserted in patients presenting with complex partial
seizures. Prolonged tracings with activation procedures,
such as sleep deprivation and careful withdrawal of
anticonvulsant agents, were performed to enhance EEG
abnormalities. Chronic depth-electrode implantation in
the right temporal lobe (for evaluation of remote epi-
leptic activity) was carried out in one patient after
excision of a right frontal AVM. Twenty patients had
focal epileptiform activities. 10 had nonspecific slow-
wave activities, and 24 had normal scalp EEG record-
ings. Intraoperative electrocorticography (ECoG) was
performed before excision of the AVM. Twenty-five

J. Neurosurg. / Volume 78 / January, 1993

patients showed interictal or ictal epileptiform activity
adjacent to or remote from the AVM’s; 11 had epilep-
tiform activity distant from the AVM’s (10 had AVM’s
at the lateral or posterior part of the temporal lobe and
one had an AVM at the anterior frontal region). Remote
epileptic activities were in the ipsilateral mesial tem-
poral region.

Surgical Procedures

If an AVM involved the speech or sensorimotor
cortical area, patients usually received local anesthesia
before undergoing intraoperative ECoG and functional
mapping of the cerebral cortex using electrical stimu-
lation. If the seizure activities involved the temporal
lobe in patients with an AVM located in the anterior,
lateral, or posterior temporal cortex or the anterior
frontal region, then acute depth electrodes were inserted
into the amygdaloid region and the anterior hippocam-
pus. Details of these operative procedures are described
elsewhere.”’ All patients had complete excision of the
AVM as confirmed by intraoperative angiography. A
postoperative cerebral angiogram was routinely per-
formed within | week after surgery and no patient had
evidence of a residual AVM. Twenty-five patients un-
derwent additional cortical excision based on preoper-
ative EEG recordings, intraoperative ECoG, and seizure
pattern; 14 had excision of the epileptic cortex adjacent
to the AVM and nine underwent mesial temporal re-
section in addition to excision of the AVM, which
involved the ipsilateral anterior or posterior lateral tem-
poral cortex. Two patients required a second surgical
procedure to excise the remote seizure focus; one had a
large AVM in the right anterior frontal region and the
other had a thrombosed AVM in the right posterior
temporal region. Both presented with complex partial
seizures caused by a remote epileptic focus in the ipsi-
lateral mesial temporal region following total excision
of the AVM.7! In these two patients, the pathology of
the mesial temporal specimen was unremarkable, with-
out evidence of vascular malformation or hemosiderin
deposit.

Illustrative Case

This 46-year-old right-handed man presented with a
27-year history of medically intractable seizures. The
seizures consisted of a sudden onset of clonic move-
ment of the right limbs, followed by unconsciousness
and generalized tonic-clonic convulsions. Postictally,
he had difficulty with speech, disturbance of the right
visual field, and left-sided headache, which lasted for 5
to 20 minutes. About two or three seizures per month
occurred despite adequate anticonvulsant therapy.

Examination. Computerized tomography with con-
trast medium infusion showed a 5-cm enhancing lesion
in the left posterior temporal region, confirmed by
cerebral angiography (Fig. | upper). Embolization with
temporary balloon occlusion of multiple feeding vessels
to the AVM was attempted, but was unsuccessful. Elec-
troencephalography showed rhythmic sharp and slow
activities in the left temporoparietal region. A carotid
Amytal (amobarbital) test disclosed a left dominant

13
Fic. 1. Preoperative cerebral arteriograms, anteroposterior (upper /efi) and lateral (upper right) views,

H.S. Yeh, J. M. Tew, Jr., and M. Gartner

showing a large arteriovenous malformation (AVM) in the left posterior temporal region. Postoperative
angiograms, anteroposterior (lower feft) and lateral (lower right) views, showing complete excision of the

AVM.

hemisphere for speech function and adequate memory
function in the right temporal lobe. The patient exhib-
ited mild impairment of receptive language skills.

Operation. A left frontotemporoparietal craniotomy
was performed with the patient under local anesthesia
on October 31, 1986. Intraoperative ECoG with surface
and acute depth electrodes showed interictal epilepti-
form activity in the left mesial temporal region (Fig. 2).
Electrical stimulation was carried out and revealed that
the speech area was superior and posterior to the AVM.
The patient was intubated and placed under general
anesthesia. The AVM and the left mesial temporal
structure showing epileptiform activity were both re-
sected.

Postoperative Course. Postoperative angiography
showed complete excision of the AVM (Fig. 1 /ower).
Pathological study revealed an AVM and gliosis of the
removed parenchyma surrounding the AVM; the me-
sial temporal specimen was unremarkable and without
evidence of vascular malformation or hemosiderin de-
posit.

The patient developed receptive dysphasia after sur-
gery. He had one episode of partial seizure with second-
ary generalization and transient right hemiplegia postic-
tally at 3 days after surgery. This episode was different
from the previous seizures and believed to be related to
postoperative swelling of the brain. He has had no
recurrent seizures since his discharge from the hospital.

14

Within 6 months, after speech rehabilitation, he had
regained speech function to the preoperative condition.

Results

Complications

There were no operative deaths among the 54 pa-
tients. Two patients developed hemiparesis, one of
whom recovered; one patient was impaired substan-
tially by upper-limb monoplegia. The AVM or seizure
foci involved the cerebral motor cortex in these three
patients. Of the other patients, one had a paresthesia
because of a small thalamic infarct, one suffered con-
duction dysphasia, and one receptive dysphasia, and
three had subtle speech dysfunction after extensive ex-
cision of the AVM and/or seizure foci in the dominant
temporal lobe. All five patients regained speech func-
tion within | year following the procedure.

Follow-Up Results

The follow-up period ranged from 18 months to 9
years (mean 4.8 years); four of 54 patients were lost to
follow-up study after initial postoperative follow-up
examination. Anticonvulsant therapy was discontinued
when patients had been seizure-free for | year, other-
wise it was continued indefinitely.

Outcome was classified as excellent if the patient
became seizure-free after surgery or had an occasional

J. Neurosurg. { Volume 78 / January, 1993
Seizure control after AVM surgery

Fic. 2. Artist's drawing of a large arteriovenous malfor-
mation in the left temporal lobe and intraoperative depth
electrode recordings, showing the interictal epileptic activity
in the left mesial temporal region (inset).

aura; good with a 90% reduction in seizure frequency;
fair with a 50% to 90% reduction in seizure frequency;
and poor with a reduction in seizure frequency of less
than 50%. Seizure control was excellent in 38 patients
(70.4%), good in 10 (18.5%), fair in five (9.3%), and
poor in one (1.8%). Two patients who underwent a
second operation to remove their remote epileptic foci
obtained excellent results. After a 5-year follow-up
study, one patient’s seizure control had improved from
good to excellent, two had deteriorated from excellent
to good, and one had deteriorated from good to fair.

Outcome of seizure control after surgery appears to
be related to patient age at seizure onset (Table 3).
Among the 30 patients aged 30 years or less at seizure
onset, 18 (60%) had excellent seizure relief after surgery:
among the 24 patients who had their first seizure at an
age greater than 30 years, 20 (83.3%) showed excellent
results.

The relationship between the location of an AVM
and the outcome of seizure control after surgery is
summarized in Table 4. All 10 patients with parietal
AVM’s, 12 of the 15 patients with frontal AVM’s, and
nine of the [7 patients with temporal AVM’s showed
excellent results in seizure control after surgery, whereas
no patient with an AVM in the rolandic or occipital
region achieved an excellent outcome. Sixteen of the
23 patients with large AVM’s, 15 of the 19 patients
with medium-sized AVM’s, and seven of the 12 patients
with small AVM’s had excellent results in seizure con-
trol after surgery. Thus, there was no significant differ-
ence in the outcome of seizure control related to size
of the AVM (Table 5).

Seizure duration and cortical excision as related to
outcome of seizure control are shown in Table 6.

J. Neurosurg. / Volume 78 / January, 1993

TABLE 3

Age al seizure onset related to outcome of seizure control after
surgery in 54 patients with cerebral AVM’s*

Age Range No. of Outcomet
(rs) Cases Excellent Good Fair Poor
11-20 18 9 4 2 0
21-30 15 9 2 3 ii
31-40 13 aI 2 0 0
41-50 6 5 1 0 0
51-60 5 4 1 0 0
totals 54 38 10 5 1

* Abbreviation: AVM = arteriovenous malformation.

{ Outcome: excellent = seizure-free or occasional auras; good =
90% reduction in seizure frequency; fair = 50% to 90% reduction in
seizure frequency; poor = < 50% reduction in seizure frequency.

TABLE 4
Location of AVM and outcome of seizure control after surgery*
. No. of Outcomet
Location C -
ases Excellent. Good Fair Poor
frontal 15 12 1 1 1
temporal 17 9 7 ] 0
rolandic 2 0 1 1 0
parietal 10 10 0 0 0
occipital 2 0 I i} 0
multilobular 8 7 0 1 0
totals 54 38 10 1

* Abbreviation: AVM = arteriovenous malformation.

t Outcome: excellent = seizure-free or occasional auras. good =
90% reduction in seizure frequency: fair = 50% to 90% reduction in
seizure frequency; poor = < 50% reduction in seizure frequency.

TABLE 5
Size of AVM and outcome of seizure control in 54 patients*
Outcomet
Sieoravm — NOOF
ass Excellent Good Fair Poor
large (> 4 cm) 23 16 4 2 1
medium (2-4 cm) 19 15 2 2 0
small (< 2 cm} 12 7 4 it 0
totals 54 38 10 5 i

* Abbreviation: AVM = arteriovenous malformation.

+ Outcome: excellent = seizure-free or occasional auras; good =
90% reduction in seizure frequency; fair = 50% to 90% reduction in
seizure frequency; poor = < 50% reduction in seizure frequency.

Eighteen (90%) of 20 patients with seizures lasting |
year or less had an excellent outcome after surgery;
based on EEG findings, five of these patients underwent
cortical excision in addition to AVM removal. Outcome
was excellent in five of seven patients with seizures
lasting 2 or 3 years, nine of 17 patients with seizures
lasting 5 to 10 years, and six of 10 patients with seizures
lasting 15 years or more; cortical excision was per-
formed in four of seven patients, eight of 17 patients,
and eight of 10 patients in the above groups, respec-

15
tively. In the group of 1! patients with remote epileptic
focus, none had seizures lasting less than 2 years. In the
25 patients undergoing cortical excision, the outcome
of seizure control was excellent in 16 (64%), good in
six (24%), and fair in three (12%) (Table 7). Twelve of
the 17 patients with temporal AVM’s and four of the
six patients with multilobular AVM’s involving the
temporal lobe underwent cortical excision.

Discussion

Natural History

Epilepsy, one of the most common clinical symp-
toms of cerebral AVM’s, may present with or without
intracranial hemorrhage. The natural history of cerebral
AVM’s, with emphasis on hemorrhage and epilepsy, is
well described.?*""*'?° However, there are insufficient
data to predict the outcome of seizure control when
epilepsy presents as the earliest and main symptom of
cerebral AVM’s. Crawford, ef al.,> concluded that
younger patients were more likely to develop epilepsy

TABLE 6

Seizure duration and cortical excision related to seizure control
afier surgery in 54 patients with cerebral AVM’s*

Duration 0. of No. with Ouicomet

of Seizure Cc . Cortical =
(yrs) aSeS Excision Excellent Good Fair Poor

th 13 3 int 2 0 0

1 7 2 7 0 0 0

2 4 2 3 ) 1 0

3 3 2 2 0 1 0

3 10 2 6 3 0 1

10 7 6 3 4 0 0

15 2 2 2 0 0 0

20 8 6 4 1 a 0

totals 54 25 38 10 5 1

* Abbreviation: AVM = arteriovenous malformation.

+ Outcome: excellent = seizure free or occasional auras; good =
90% reduction in seizure frequency; fair = 50% to 90% reduction in
seizure frequency: poor = < 50% reduction in seizure frequency.

TABLE 7

Location ef AVM and cortical excision related to seizure
control after surgery in 25 patients*

No. of Cases Outcomet
Location ain o_O
Excision Excellent. Good Fair Poor
frontal 4 2 2 0 0
lernporal 12 7 4 1 0
rolandic i 0 0 1 0
parietal 2 2 0 0 0
occipital 0 0 0 0 0
multilobular 6 5 0 1 0
totals 25 16 6 3 0

* Abbreviation: AVM = arteriovenous malformation.

T Outcome: excellent = seizure-free or occasional auras; good =
90% reduction in seizure frequency; fair = 50% reduction in seizure
frequency; poor = < 50% reduction in seizure frequency.

16

H. S. Yeh, J. M. Tew, Jr., and M. Gartner

during the follow-up period and that the risk of devel-
oping epilepsy was greatest if the AVM involved the
temporal lobe. We postulate that the mechanisms of
epilepsy associated with cerebral AVM’s are the follow-
ing: focal cerebral ischemia secondary to arteriovenous
shunting, gliosis of the brain surrounding the AVM
caused by a previously subclinical hemorrhage, or he-
mosiderin deposits and a secondary epileptogenesis in
the temporal lobe.”

Electroencephalographic Findings

It is not unusual that the routine EEG findings are
normal in patients with cerebral AVM’s, especially in
those with seizures of short duration. Implantation of
subdural or chronic depth electrodes for monitoring
seizures is not considered unless the primary vascular
lesion is removed because of the high risk of bleeding.
Therefore, we relied mostly on ECoG during surgery to
define the extent of cortical excision.

Secondary Epileptogenesis

Secondary epileptogenesis in humans has been de-
scribed in patients with epilepsy caused by brain tumor
and cerebrovascular malformations.'*!*?' In experi-
mental animals, development of a secondary independ-
ent epileptic focus contralateral or homolateral to the
primary lesion is well documented; the time required
for the genesis of a secondary epileptic focus varied
with the animals studied.''”'® About 21% to 36% of
patients with epilepsy caused by brain tumors demon-
strate secondary epileptogenesis; approximately one-
third of these patients develop secondary epileptic foci
and continuing clinical seizures after removal of the
primary tumor lesion.'*"'5 Morrell’? postulated that sec-
ondary epileptogenesis exists in both reversible and
irreversible forms. The reversible form develops first
and is followed by the irreversible form, which is char-
acterized by seizures of longer duration and greater
frequency.'> Eleven patients in our series had epileptic
foci distant from the primary AVM, with 10 AVM’s in
the lateral or posterior part of the temporal lobe and
one in the ipsilateral anterior frontal region. All remote
epileptic foci were in the ipsilateral mesial temporal
region. Two patients underwent a second procedure to
resect the ipsilateral mesial temporal structure because
active epileptic discharges continued after the first sur-
gery to excise the AVM, which was at the posterior
temporal or the anterior frontal region.”' In these two
patients, seizures persisted for 9 and 20 years. Clearly,
secondary epileptogenesis may occur in humans who
have had cerebral AVM’s presenting with chronic epi-
epsy.

Factors Affecting the Outcome of Seizure Control

Age of Seizure Onset. In this series, there appeared
to be a difference in outcome if seizures began after 30
years of age. Excellent seizure control was demonstrated
in 60% of patients whose seizures occurred at age 30
years or less and in 83.3% of those whose seizures
occurred after age 31 years. Similarly, in patients with
epilepsy caused by brain tumors, the likelihood of
secondary epileptogenesis appeared to depend on age

J. Neurosurg. / Volume 78 / January, 1993
Seizure control after AVM surgery

at epilepsy onset.'? In an animal study, a decreased
susceptibility to kindling was found in aged animals.*

Duration of Seizures. In this series, patients had
favorable outcomes when seizures were of shorter du-
ration. Excellent postoperative results were achieved in
90% of patients with a seizure history of | year or less;
only 25% of these patients underwent cortical excision.
The longer the duration of seizures, the more likely
seizure control is to be poor, despite higher rates of
cortical excision. Apparently, the need for cortical ex-
cision is more often based on EEG studies in patients
with longer duration of seizures, but seizure control is
still no better than in patients experiencing seizures of
shorter duration.

Location of Lesion. In our series, 12 (71%) of the
17 patients with temporal AVM’s had associated de-
monstrable epileptic foci. Eleven patients (one with a
fronta!t AVM and 10 with temporal AVM’s) had remote
epileptic foci in the ipsilateral mesial temporal struc-
ture. Goddard, et ai." demonstrated that the limbic
system structure in rats, especially the amygdala, is the
most sensitive to electrical kindling. In a study of
secondary epileptogenesis in patients with epilepsy
caused by brain tumors, temporal lobe tumors were
associated with the highest incidence of independent
epileptic discharges at the contralateral or the ipsilateral
side.'* Patients with AVM’s in the rolandic area had
less favorable outcomes for seizure control because of
a limitation of cortical excision when epileptic dis-
charges involve the motor cortex in the precentral gyrus.

Cortical Excision. Cortical excision remains the
standard surgical procedure for epilepsy if the epilep-
togenic focus is localized and accessible. Twenty-five
(46%) of the 54 patients in our series underwent cortical
excision in addition to AVM removal. Results of sei-
zure contro! after surgical management of AVM are
much better in our first series” when compared with
others.”'* Cortical excision appeared to be indicated
more often in patients with seizures of earlier onset and
longer duration. The role of other therapeutic modali-
ties in seizure outcome in patients with cerebral AVM’s
is not clear. Fournier, ef a/.,° reported that five (24%)
of 21 patients experienced a significant reduction in
seizure frequency after complete or partial embolization
of cerebral AVM’s. Lunsford, e¢ a/.,'' noted that of 43
patients with seizures before undergoing gamma-knife
radiosurgery, 22 (51%) showed improved seizure con-
trol after treatment when evaluated | to 3 years later.
Heikkinen, ef al.,'" described 16 (55%) of 29 patients
with cerebral AVM’s who became seizure-free after the
AVM's were treated by stereotactic proton-beam irra-
diation; they suggested that ionizing radiation could
inhibit epileptic activity around the AVM. We are
concerned that radiotherapy of the AVM has no effect
on epileptic foci occurring at a distance from the AVM.
The possible effect of focal radionecrosis on seizure
control in Jong-term follow-up monitoring is uncertain.

Conclusions

The results of this prospective study of 54 patients
with cerebral AVM’s presenting with epilepsy showed

J. Neurosurg. / Volume 78 / January, 1993

that cortical excision in selected patients improved the
outcome of seizure control. The presentation of de-
monstrable epileptic foci and their intractability ap-
peared to correlate with age at seizure onset, duration
of seizures, and location of lesions. Secondary epilep-
togenesis does exist in humans with cerebral AVM’s;
however, the question remains as to how often it occurs
and how seriously it affects the outcome of seizure
control. Early surgery of cerebral AVM’s in young
patients presenting with epilepsy should be considered.
Further investigations for localizing epileptic foci are
warranted in patients who continuously suffer from
intractable epilepsy after surgical excision of cerebral
AVM’s.

Acknowledgments

The authors thank Nancy McMahon for material collec-
tion, Sam Scott, Ph.D., and Scott Michaels, Ph.D., for statis-
tical analysis, Alexis Rostoker for manuscript preparation,
and Mary Kemper for editorial assistance.

References

i. Ajmone Marsan C: Unitary analysis of “projected” epi-
leptiform discharges. Electroencephalogr Clin Neurophys-
jol 15:197-208, 1963

2. Crawford PM, West CR, Chadwich DW, et al: Arterio-
venous malformations of the brain: natural history in
unoperated patients. J Neurol Neurosurg Psychiatry
49:1-10, 1986

3. Crawford PM, West CR, Shaw MDM, et al: Cerebral
arteriovenous malformations and epilepsy: factors in the
development of epilepsy. Epilepsia 27:270-275, 1986

4, de Toledo-Morrell L, Morrell F, Fleming S: Age-depend-
ent deficits in spatial memory are related to impaired
hippocampal kindling. Behav Neurosci 98:902-907, 1984

5. Forster DMC, Steiner L, Hakanson $: Arteriovenous
malformations of the brain. A long-term clinical study. J
Neurosurg 37:562-570, 1972

6. Fournier D, TerBrugge KG, Willinsky R, et al: Endovas-
cular treatment of intracerebral arteriovenous malfor-
mations: experience in 49 cases. J Neurosurg 75:228-233,
1991

7. Fults D, Kelly DL Jr: Natural history of arteriovenous
malformations of the brain. A clinical study. Neurosur-
gery 15:658-662, 1984

8. Goddard GV, McIntyre DC, Leech CK: A permanent
change in brain function resulting from daily electrical
stimulation. Exp Neurol 25:295-330, 1969

9, Graf CJ, Perret GE, Torner JC: Bleeding from cerebrat
arteriovenous malformations as part of their natural his-
tory. J Neurosurg 58:331-337, 1983

10, Heikkinen ER, Konnov B, Melnikov L, et al: Relief of
epilepsy by radiosurgery of cerebral arteriovenous mal-
formations. Stereotact Funct Neurosurg 53:157-166, 1989

11, Lunsford LD, Kondziolka D, Flickinger JC, et al: Ste-
reotactic radiosurgery for arteriovenous malformations
of the brain. J Neurosurg 75:5 12-524, 1991

12. Michelsen WJ; Natural history and pathophysiology of ar-
teriovenous malformations. Clin Neurosurg 26:307-313,
1

13, Morrell F: Secondary epileptogenesis in man. Arch Neu-
rol 42:3 18-335, 1985

14, Morrell F, Rasmussen T, Gloor P, et al: Secondary epi-
leptogenic foci in patients with verified temporal lobe

17
15,

18

tumor. Electroencephalogr Clin Neurophysiol 56:26, 1983
(Abstract)

Morrell F, Wada JA, Engel J Jr: Potential relevance of
kindling and secondary epileptogenesis to the considera-
tion of surgical treatment of epilepsy, in Engel J Jr (ed):
Surgical Treatment of Epilepsies. New York: Raven
Press, 1987, pp 699-707

. Parkinson D, Bachers G: Arteriovenous malformations.

Summary of 100 consecutive supratentorial cases. J Neu-
rosurg 53:285-299, 1980

. Wilder BJ, King RL, Schmidt RP: Comparative study of

secondary epileptogenesis. Epilepsia 9:275-289, 1968

. Wilder BJ, Morrell F: Secondary epileptogenesis in the

frog forebrain. Neurology 17:1041-105t. 1967

. Wilkins RH: Natural history of intracranial vascular mal-

formations. A review. Neurosurgery 16:421-430, 1985

20.

21.

22.

pa
of
05

H.S. Yeh, J. M. Tew, Jr., and M. Gartner

Yeh HS, Kashiwagi S. Tew JM Jr, et al: Surgical manage-
ment of epilepsy associated with cerebral arteriovenous
malformations. J Neurosurg 72:216-223, 1990

Yeh HS, Privitera MD: Secondary epileptogenesis in ce-
rebral arteriovenous malformations. Arch Neurol 48:
1122-1124, 1991

Yeh HS, Tew JM Jr: Management of arteriovenous mal-
formations of the brain. Contemp Neurosurg 9(26):1-8,
1988

Manuscript received January 10, 1992.

Accepted in final form July 7, 1992.

Address reprint requests to: Hwa-shain Yeh, M.D., De-
rtment of Neurosurgery, University of Cincinnati College
Medicine, 231 Bethesda Avenue, Cincinnati, Ohio 45267-
15.

J. Neurosurg. / Volume 78 / January, 1993
uy ¥
rN
OX

~
en
x