Epilepsia. 3I(Suppl. 3):S37-S44, 1990

Raven Press, Ltd.. New York
0 1990 International League Against Epilepsy

Epilepsy and Brain Tumors: Implications for Treatment
Gregory D. Cascino
Department of Neurology, Mayo Medical School, Rochester, Minnesota, U.S.A.

~

Summary:Primary intraparenchymal tumors of the brain

~

tumors that are deep-seated lesions or involve functional
cerebral cortex. Computer-assisted stereotactic surgical
procedures have been developed for biopsy and resection
of intra-axial brain-mass lesions. Stereotactic tumor resection may allow pathological determination of intracranial lesions and produce a worthwhile reduction in seizure
activity in some patients with intractable partial epilepsy.
Key Words: Epilepsy-Seizures-Brain
neoplasmGliomas-Neurological surgery-Stereotaxic technics.

are important etiologic factors in partial or focal epilepsy.
Indolent low-grade gliomas may be associated with a
long-standingseizure disorder refractory to medical treatment. Surgical resection of the neoplasm and the epileptogenic area may render patients seizure-free. Removal
of the tumor alone may also be associated with an excellent survival rate and surgical outcome. Conventional
neurosurgical procedures are restricted in patients with

The occurrence of seizures and epilepsy in the
presence of primary brain tumors has been observed
for over one century (Jackson, 1882,1958; Ketz,
1974). Hughlings Jackson determined that the ictal
phenomena, including seizure type, may indicate
the localization of the neoplastic lesion (Jackson,
1882,1958). He emphasized that a chronic seizure
disorder could be symptomatic of a glioma in a patient with partial and secondarily generalized seizures (Jackson, 1882). Jackson made the important
observation that the epilepsy could be the initial and
only clinical manifestation of a brain tumor (Jackson, 1882,1958). The severity of the seizure disorder
and the type of seizure, however, could not predict
the neuropathology associated with the epilepsy
(Jackson, 1882).
The advent of neurosurgery in treating epilepsy
and central nervous system tumors has confirmed
these early observations. Seizures occur in approximately 50% of patients with intracerebral neoplasms (Rasmussen, 1975). Neuroectodermal tumors are the most common primary parenchymal
tumors of the central nervous system, and the majority are gliomas (Shapiro and Shapiro, 1986; Kornblith et al., 1987). The incidence of seizures and
epilepsy associated with primary brain tumors is related to the tumor pathology and cortical localiza-

tion (Lund, 1952; Pedield et al., 1940; Ketz, 1974,
Le Blanc and Rasmussen, 1974). “Slow-growing”
low-grade or well-differentiated gliomas (Kernohan
and Sayre grades I and 11)-astrocytoma, oligodendroglioma, and oligoastrocytoma-are the most epileptogenic lesions (Pedield et al., 1940; Lund, 1952;
Rasmussen and Blundell, 1959; Ketz, 1974; Le
Blanc and Rasmussen, 1974; Rasmussen, 1975). The
ganglioglioma, a neural-series tumor with an astrocytic component, and the dysembryoplastic neuroepithelial tumor, composed of glial cells and neurons, are also important neoplastic determinants of
a chronic seizure disorder (Kornblith et al., 1987;
Daumas-Duport et al., 1988). Tumors located proximal to the centroparietal region more frequently are
associated with epilepsy (Penfeld et al., 1940; Ketz,
1974; Le Blanc and Rasmussen, 1974).
Seizures are the most common presenting s y m p
tom in patients with low-grade tumors. There is a
direct correlation between the chronicity of the
growth of the neoplasm and the incidence of seizures. In the Montreal series of 230 patients with
gfiomas, seizures occurred in 70% of those with astrocytomas, in 92% of those with oligodendrogliomas, and in 37% of those with glioblastomas (Penfield et al., 1940). These figures are slightly higher
than those in other reported studies, a difference
reflecting the referral of patients with seizure disorders (Ketz, 1974). The incidence of epilepsy is
lower in patients with glioblastoma multiforme, perhaps because of the shorter duration of the disease

Address correspondenceand reprint requests to Dr. G . D. Cascino at Department of Neurology, Mayo Medical School, Rochester, MN 55905, U.S.A.

s37

S38

G . D . CASCINO

(Penfield et al., 1940; Ketz, 1974). Patients with malignant gliomas, i.e., anaplastic astrocytoma and
glioblastoma multiforme, are more likely to present
with focal or generalized cerebral symptomsrelated
to a space-occupying lesion (Shapiro and Shapiro.
1986).

The occurrence of brain tumors in patients with
epilepsy has varied with the age of the patients,
type(s) of seizures, and the nature of the patient
population studied (Ketz, 1974). The most commonly identified neoplasm in patients with intractable epilepsy is the low-grade glioma (Le Blanc and
Rasmussen, 1974). In patients with partial epilepsy
refractory to medical treatment who undergo surgical ablative procedures, the frequency of tumors
is approximately 10-20% (Penfield et al, 1940; Le
Blanc and Rasmussen, 1974; Babb and Brown,
1987). Computed tomographic (CT) studies in patients with epilepsy have demonstrated a neoplasm
in 1625% of patients with onset of seizures after
the age of 40 years and in fewer than 5% of children
(Shorvon, 1987). During presurgical evaluation of
190 patients in one study of intractable partial epilepsy, 15% were found to have an unsuspected intracranial mass lesion and 10% a neoplasm (Spencer
et al., 1984). In the Rochester series of 516 patients
with epilepsy, which was not restricted to patients
with partial seizure disorders, the incidence of neoplastic lesions was 4.1% (Hauser and Kurland,
1975). Tumors occur more frequently in patients
with an onset of epilepsy after the age of 20 years;
however, in children and in adolescents, brain tumors remain an important cause of intractable partial epilepsy (Blume et d . , 1982; Suarez et al., 1986;
Drake et al., 1987).
The mechanism(s) of epileptogenesis associated
with brain tumors is unknown (Ketz, 1974). Hughlings Jackson observed that the pathogenesis of epilepsy in patients with tumors was related to the presence of a slow-growingtumor in the gray matter and
the “sudden excessive and temporary discharge of
nerve cells,” producing seizure activity (Jackson,
1882,1958; Ketz, 1974). Penfield suggested that impaired vascularization of the surrounding cerebral
cortex may produce hypoxic-ischemic neuronal
changes (Ketz, 1974). Direct “irritation” of the cortex by a tumor has also been proposed as the etiology of seizure activity. In the symptomatic partial
epilepsies, a focal structural abnormality is intimately associated with the epileptogenic area, the
site of seizure onset (Gloor, 1987; Engel. 1989). Specifc morphologic neuronal alterations related to a
focal brain lesion, e.g., tumors, have been observed
that are potentially epileptogenic (Engel, 1989).
Changes in y-aminobutyric acid (GABA)-ergic synEpilcpsia, Vol. 31. Suppl. 3, 1990

apses, the N-methyl-paspartic acid (NMDA) receptor-cation channel, or axonal calcium or chloride channels may be associated with the enhanced
excitatory andor attenuated inhibitory influences
that may facilitate epileptogenesis (Engel, 1989).
Brain tumors may produce “denervation hypersensitivity” related to a partial isolation and transection
of a region of cerebral cortex (Echlin, 1959; Engel,
1989). There is conflicting evidence concerning the
possible importance of hereditary factors in the development of epilepsy in patients with tumors
(Ketz, 1974).
A long-standing partial or localization-related seizure disorder may be related to an indolent lowgrade glioma with a clinical course indistinguishable
from that of a non-neoplastic process (Rasmussen
and Blundel, 1959; Le Blanc and Rasmussen, 1974;
Rasmussen, 1975; Goldring et al., 1986). Epilepsy
occurring in the presence of a brain tumor may be
associated with a simple partial seizure activity,
“changing” neurological examinations, or persistent focal nonepileptiform changes in the electroencephalogram (EEG) (Blume et al.. 1982; Spencer
et al., 1984; Dam et al.. 1985). However. available
evidence conflicts regarding the importance of these
variables in identifying a neoplasm-related seizure
disorder (Le Blanc and Rasmussen. 1974; Blume et
al., 1982; Spencer et al., 1984). The most important
parameter in the diagnostic evaluation to differentiate a neoplastic from a non-neoplastic etiology in
patients with partial epilepsy is the neuroimaging
study, CT scan, or magnetic resonance imaging
(MRI) (Goldring et al.. 1986; Sperling et al.. 1987).
MRI is particularly sensitive in the detection of
“foreign-tissue” lesions. e.g., tumors and vascular
malformations (Sperling et al., 1987). Electrophysiologic studies are necessary to demonstrate an association between the neuroimaging lesion and the
epileptogenic area (Engel, 1987~).MRI is superior
to CT in imaging gliomas and therefore is the procedure of choice (Sperling et al.. 1987).
SURGICAL THERAPY FOR EPILEPSY
ASSOCIATED WITH BRAIN TUMORS

Surgical therapy is performed in the management
of patients with epilepsy and low-grade gliomas to
resect the tumor and to reduce seizure activity without associated neurologic morbidity. The efficacy
of surgical treatment of partial epilepsy associated
with neoplastic lesions has previously been documented (Rasmussen. 1975; Goldring et al.. 1986:
Drake et al., 1987). Successful treatment of the seizure disorder may not only render the patients seizure-free but also permit a reduction in antiepileptic

EPILEPSY AND BRAIN TUMORS
drug toxicity and improve psychological and cognitive performance. The treatment of patients with
these lesions and without epilepsy will not be considered. Surgical management of malignant gliomas
is performed to establish a tissue diagnosis and to
reduce the bulk of the tumor and will also not be
included in this discussion.
Penfield et al. (1940) observed that the surgical
result, i.e., seizure outcome, in patients with tumors
may be as good as the outcome in those with focal
atrophic lesions if the tumor is encapsulated and can
be extirpated. The recurrence of the neoplasm may
be associated with partial seizure activity after a
seizure-free interval. Surgical procedures have included resection of the neoplastic lesion and the surrounding epileptogenic area. Penfield and Jasper
(1954) suggested that surgical removal of a structural abnormality alone may not produce a reduction in seizure activity and that a second procedure
may be required to resect the epileptogenic cortex.
However, it has been demonstrated that excision of
a low-grade glioma without resection of the epileptogenic area may have a favorable effect on seizure
activity (Goldring et al., 1986; Hirsch et al., 1989).
Conventional neurosurgical procedures may be restricted in patients who have tumors that are deepseated lesions or involve functional cerebral cortex
(Apuzzo and Sabshin, 1983; Kelly, 1986,1988; Kelly
et al., 1987). Computer-assisted stereotactic surgical resection of intra-axial brain-mass lesions has
been developed (Apuzzo and Sabshin, 1983; Kelly,
1986,1988; Kelly et al., 1987; Wilden and Kelly,
1987). The stereotactic technique is superior to conventional neurosurgical procedures for the precise
localization of intra-axial abnormalities (Kelly,
1988). Stereotactic surgical resections can be used
to biopsy and resect intracranial lesions that involve
functional cerebral cortex with a low morbidity
(Kelly, 1988). Initial observations after stereotactic
resection of primary brain tumors in patients with
epilepsy suggested a reduction in seizure activity
(Kelly, 1988).
Volumetric stereotactic surgical resection of
intra-axial brain-mass lesions has been performed
in more than 300 patients at the Mayo Clinic. A
review of 226 consecutive resections between 1984
and 1987 showed the 226 lesions to include 112 glial
tumors and 70 nonglial tumors; 36 of the 226 patients
had medically refractory partial seizure disorders;
all were seizure-free at 3 months postoperatively
(Kelly, 1988). The methodology for the volumetric
stereotactic computer-assisted procedures has been
reported previously (Kelly, 1986,1988; Kelly et al.,
1987). The database acquisition and the actual surgical procedure are performed on two separate

s39

days. An MRI- and CT-compatible headframe is
used, and a CT scan, MRI, and a digital subtraction
angiogram are performed on all patients. The initial
data acquisition is used to determine a surgical avascular trajectory for resection of the lesion. The archival data tapes from the CT scan, MRI, and angiogram are then run into an operating-room
computer system. At the time of the surgical procedure, the patient’s head is again placed in the stereotactic headframe. The surgical technique depends
on whether the lesion is deep-seated or is a superficial cortical abnormality. Deep intracranial lesions
are approached through a 3.8-cm cranial trephine.
A stereotactically directed, cylindrical retractor 2
cm in diameter is used to maintain the exposure to
the lesion. The position of the circular trephine is
set by the stereotactic headframe. The computer
displays the tumor-slice image within a circle that
represents the position and configuration of the cranial trephine. The configuration of CT- and MRIdefined tumor slices is also superimposed on the
surgical field to indicate the location, configuration,
and extent of the subcortical lesion at successive
depth levels. Display information is shown on a
video monitor in the operating room. A carbon dioxide laser is used to separate the lesion from the surrounding brain tissue. The tumor is then removed
by laser vaporization.
A retrospective study of 45 consecutive stereotactic resections of intra-axial lesions in patients
with partial epilepsy was performed (Cascino et al.,
1989; Cascino et al., 1990). Thirty patients had medically refractory seizure disorders. The most common pathologic alterations observed were vascular
malformations (20 lesions) and glial neoplasms (16
lesions). The lesion localization included the postcentral gyrus (N = 9), precentral gyrus ( N = 8).
deep-seated left posterior temporal region ( N = 6),
left fronto-operculum ( N = I), and occipital cortex
(N = 2), sites that may have precluded the use of
standard craniotomy with cortical resections. Seizure outcome postoperatively was classified as
class I (seizure-free or auras only without loss of
consciousness or atypical seizures during antiepileptic drug discontinuation), class I1 [rare seizures
(1-2 per year) since surgery or initially seizure-free
and now rare seizure activity], class 111 (continued
partial seizures with an 80% reduction in seizure
activity), and class IV (no improvement, although
some reduction in seizures may have occurred)
(Engel, 19876). Forty-two of the 45 patients and 27
of the 30 refractory to medical treatment had at least
an 80% reduction in seizure activity during a mean
of 20 months postoperatively. In the medically refractory patients, 18 of 20 patients and 13 of 15 paEpilepsia. Vd.31. Suppl. 3. 1990

G . D . CASCINO

S40

tients, who were followed for at least 1 year (mean
of 30 months) and 2 years (mean of 34 months),
respectively, had favorable surgical outcomes (i.e.,
class I, 11, or 111). Thirteen of the 20 patients and 9
of the I5 patients followed for at least 1 year and 2
years, respectively, had a class I surgical outcome.
The brain tumors included grade I astrocytoma
(N = 3, grade I oligodendroglioma ( N = 4).
grade I1 oligodendroglioma (N = 4), grade I
oligoastrocytoma( N = I), and grade I ganglioglioma
(N = 2). The mean age of the 16 patients with neoplastic lesions was 20 years (range of 3-43 years).
In the 16 patients with glial neoplasms, 11 had medically refractory seizure disorders. All five patients
whose seizures had been medically controlled preoperatively were seizure-free after lesionectomy for
a mean follow-up period of 19 months postoperatively. The results of the stereotactic tumor resections in the 11 patients at a mean follow-up period
of 25 months (range of 6-46 months) were as follows: class I, 9 patients; class 111, 1 patient; and
class IV, 1 patient. The nine patients in class I included five who were seizure-free while receiving
medication, three who had a single generalized seizure associated with discontinuation of medication,
and one who had auras only. The mean duration of
follow-upfor the two patients who were in class 111
and class IV was 10 months and 24 months, respectively.
Among the entire group of 45 patients, the o p
erative morbidity included a language or motor deficit in three patients. One of these patients, after
resection of a left mesial posterior temporal vascular
malformation, developed an anomic aphasia that
gradually improved, with a nearly complete recovery noted at 1 year postoperatively; a second acquired a transient right hemiparesis in the immediate
postoperative period related to a left parietal vascular malformation resection; and a third underwent
a surgical procedure that resected a structural lesion
from the right motor cortex. No diagnostic abnormality was noted in the third patient. A left monoparesis involving the upper extremity occurred;
the only persistent deficit was a reduction in rapid
altering movements in the left hand. No morbidity
was identified in the biopsy and resection of the 16
gliomas. No surgically related infections, intracranial hemorrhages, or deaths occurred.
ILLUSTRATIVE CASES

case I
A 24-year-old right-handed man had a partial seizure disorder of 10 years' duration. The patient initially had secondarily generalized seizures. Four
Epilepsia, Vol. 31. Suppl. 3 , 1990

FIG. 1. An MRI in the coronal plane obtained with u8a of a
long repetition timdecho time pulse sequence revealing an
abnormality in the right parietal region. (Note that the right
cerebral hemisphere is on the left side of the photograph.)

years before evaluation, the patient began having
"petit mal" seizures, which included verbal automatisms in which he would speak fluently but in a
confused manner. The patient's condition was refractory to maximally tolerated monotherapy and
combination therapy with antiepileptic drugs. Prolonged video EEG monitoring revealed intermittent
right frontocentrotemporal spike discharges with
spike and slow-wave complexes. Multiple clinical
and electrographic seizures were recorded associated with an ictal onset in the right frontoparietal
region. MRI revealed a "small bulge" in the right
parietal lobe consistent with a neoplasm (Fig. I). A
computer-assisted stereotactic craniotomy was performed with the resection of the right parietal lesion.
A grade I pleomorphic xanthoastrocytoma was
demonstrated. The preresection electrocorticogram
revealed active spike activity anterior and posterior
to the lesion, with the anterior spiking tending to
lead the posterior spiking. An EEG performed 4
months after resection recorded rare right temporal
spike activity during sleep. Prior to surgery, the patient was having four to five complex partial seizures per month. No neurologic morbidity was associated with the patient's surgical procedure.
Currently, approximately 6 months postoperatively. the patient i s seizure-free and asymptomatic
while receiving a reduced program of antiepileptic
medication.

EPiLEPS Y AND BRAIN TUMORS

s4i

Case II
A 12-year-oldleft-handed boy had a medically refractory partial seizuredisorder of 5 years’ duration.
The patient had generalized tonic-clonic seizures
and complex partial seizures. Secondarily generalized events began with focal tonic-clonic activity
on the left side. The patient was also having episodes of amnestic automatisms every 3 days. prolonged video EEG monitoring noted prominent right
frontocentral spike activity. MRI revealed a right
posterior frontal opercular region lesion (Fig. 2).
Computer-assisted stereotactic resection was performed and revealed a grade I oligoastrocytoma.
The preresection electrocorticogram revealed superior temporal and inferior central spike discharges. The postresection electrocorticogram
noted reduced superior temporal suprasylvian
spikes. Sleep and awake EEGs revealed no epilep
tiform activity 4 months after surgery. The patient
is seizure-free and asymptomatic 3 years after surgery. The MRI has not revealed evidence of recurrence of the neoplasm. The patient is still receiving
maintenance monotherapy with phenytoin. No morbidity was associated with the surgical procedure.

case III
A 6-year-old child with a seizure disorder of 18
months’ duration was observed to have frequent
focal tonic-clonic seizures involving the right lower
extremity. No definite impairment of consciousness
was noted. The patient was demonstrated to have
a medically refractory seizure disorder. Neurologjc
examination disclosed no definite abnormality. The
ictal EEG onset was in the left centroparietalregion.
MRI revealed a lesion in the left frontoparietal parasagittal region (Fig. 3). A computer-assisted stereotactic biopsy of the left motor cortex lesion was
performed. Pathologic examination revealed a
grade I pilocytic astrocytoma. No neurologic morbidity was noted postoperatively. At 1 year after
surgery, the patient is seizure-free with no obvious
recurrence of the tumor but is observed to have a
developmental delay. Sleep and awake EEGs were
unremarkable at this time. Antiepileptic medication
has been continued.
DISCUSSION

FIG. 2. (A) An MRI in the axial plane revealing an area of
increased intensity in the right fronto-opercular region. (a)
An MRI in the coronal plane. (Note that the riaht cerebral
hemisphere is on the leit side of the photograph.) Both images were obtained with the use of a long repetition
timdecho time pulse sequence.

Computer-assisted stereotactic resection of lowgrade gliomas in patients with epilepsy may be preferred to a standard operative resection of the tumor
and epileptogenicarea in select circumstances,e.g.,
on the basis of the results of a sodium amob&iM
study Or On the 1ocdiZatiOn Of the lesion in fUnCtional cerebral cortex. The data acquisition prior to
the stereotactic procedure may allow for be least
Epilepsio. Vd.31. Supp!. 3, 1990

S42

G . D . CASCINO

FIG. 3. (A) An MRI in the axial plane revealing an area of
increased intensity in the left frontoparietal parasagittal region. (B) An MRI in the coronal plane. (Note that the left cerebral hemisphere is on the right side of the photograph.)Both
images were obtained with the use of a long repetition
timdecho time pulse sequence.

possible disruption of normal brain tissue. The computer display images may provide information regarding the boundaries of the structural abnormality
and normal brain tissue that may not be apparent at
the time of gross resection (Kelly, 1988). Potentially, stereotactic resection may result in less neurologic morbidity than that associated with the standard neurosurgical procedures.
Epilepsia. Vol. 31, Suppl. 3. 1990

Excision of a low-grade glioma alone, without resection of the epileptogenic cortex, may be associated with a favorable seizure outcome (Goldring
et al., 1986; Hirsch et al., 1989). Goldring et al.
(1986) evaluated the surgical outcome after temporal lobe tumor excision, usually a low-grade astrocytoma, in 20 patients with epilepsy. Only one
patient continued to have postoperative seizures,
although two patients had seizures early in the postoperative period. The mean follow-up period was
2.5 years (range of 3 months to 10 years). The authors emphasized the presence of a chronic seizure
disorder (the mean interval from seizure onset to
surgery was 1 1 years) and an isolated lesion on CT
examination in these patients. The type of seizure
and the lack of response to medication did not seem
to identify the nature of the epileptogenic lesion.
Patients who had a simple excision of a temporal
lobe tumor were compared with patients who had
an anterior temporal lobectomy that included the
tumor and hippocampus. During a comparable follow-up period. (> 1.5 years; range of 1.5-10 years),
1 of 12 patients with the simple excision and 2 of 7
patients with a lobectomy continued to have seizures after surgery. Hirsch et al. (1989) evaluated
the response to tumor resection alone in 42 children
with low-grade gliomas. The incidence of epilepsy
preoperatively was 78%. Tumor removal alone was
performed without resection of the epileptogenic
cortex. The incidence of epilepsy was 1Wo after surgery. The follow-up period was variable (range of 6
months to 17 years).
Resection or alteration of the epileptogenic cortex
surrounding the lesion is a potential mechanism
whereby stereotactic tumor excision may reduce
seizure activity. Disruption of the cerebral cortex
will occur inevitably with the stereotactic procedure. Even incomplete resection of the epileptogenic area may be associated with a significant
reduction in seizure activity (Ojemann, 1985).
Enhancement of the pharmacologic responsiveness
of medically refractory seizures after tumor removal
cannot be excluded. It has been proposed that lesionectomy “restores a mechanism that limits the
spread of seizures and, thus, the clinical manifestations of epilepsy” (Hirsch et al., 1989). However,
tumor excision alone clearly does not abate seizure
activity in all patients (Cascino et al., 1989; Cascino
et al., 1990). The failure of this procedure may relate
to the remoteness of the epileptogenic area or to its
noncontiguity with the tumor. The stereotactic procedure may also be unable to resect a diffuse or
infiltrating glioma. The possibility of dual pathology, e.g., coexistent hippocampal sclerosis, cannot
be excluded (Babb and Brown, 1987; Drake et al.,

EPILEPSY A N D BRAIN TUMORS
1987). However, the most likely explanation for the
continued partial seizure activity postoperatively is
the inadequate resection of the epileptogenic area
(Gloor, 1987).
This discussion represents an initial observation
regarding stereotactic lesionectomy in patients with
partial epilepsy and neoplastic lesions. An ideal
study would have included a prospective analysis
of all patients who have had a long-term follow-up
period, e.g., a minimum of 2 years (Engel, 19876).
The possibility of false reporting, i.e., of increasing
favorable outcome regarding seizure activity, is difficult to exclude in a retrospective analysis (Engel,
1987b). The follow-up period of the patient group is
variable, and long-term control of seizures cannot
be assumed. Certainly the most meaningful data for
the patients with intractable seizure disorders are
the identification of patients who are seizure-freeor
nearly seizure-free who have been followed for at
least 1 or 2 years postoperatively (Engel, 1987b). In
the 16 patients with neoplastic lesions, 14 had class
1 surgical outcomes after stereotactic lesionectomy
during a short-term period of observation. Importantly, stereotactic lesionectomy may be associated
with reduced seizure activity without associated
neurologic morbidity even for the resection of lesions involving functional cerebral cortex.
Potentially, the surgical outcome may have been
more favorable after resection of the tumor and the
epileptogenicarea. In our series, 2 of the 1 1 patients
with medically refractory seizure disorders and
brain tumors had persistent partial seizure activity
after lesionectomy. Four of the class I patients had
atypical single seizure episodes or persistent auras.
Only 5 of the 1 1 patients with neoplastic lesions
were seizure-free, and all of the patients continued
to receive antiepileptic medication. It is anticipated
that the long-term follow-up of these patients would
demonstrate an increased number with recurrent
seizure activity. One patient, after a stereotactic resection of a lateral temporal oligodendroglioma,was
seizure-free for 9 months before partial seizure activity recurred. An anterior temporal lobectomy
was subsequently performed; intraoperative depth
recordings revealed active hippocampal spiking
ipsilateral to the lesion. A comparison of surgical
techniques is difficult because the location of the
lesion in some instances may have precluded a standard operative resection.
Potentially, the stereotactic lesionectomy may
complement other surgical techniques in the management of patients with epilepsy. The ultimate aim
for the surgical treatment of epilepsy is not only to
reduce seizure activity but also to avoid potential
morbidity associated with a therapeutic intervention

s43

(Ojemann and Dodrill, 1985). Functional extraoperative and intraoperative cortical stimulation
studies, however, may also have allowed resection
of these tumors and the epileptogenic area with a
favorable seizure outcome and without neurologic
morbidity (Ojemann, 1983; Goldring and Gregorie,
1984; Lesser et al., 1984). The evaluation of the surgical outcome must include assessment of acquired
postoperative neurologic deficits and the effect on
seizure activity (Ojemann and Dodrill, 1985).
There are important limitations regarding the role
of stereotactic neurosurgery in the management of
partial epilepsy. Neuroimaging studies may identify
a potentially epileptogenic abnormality in the rninority of patients with partial epilepsy. The most
commonly identified pathologies associated with intractable partial epilepsy, i.e., mesial temporal sclerosis and cortical gliosis, may not be associated with
a neuroimaging abnormality (Sperling et al., 1987).
Presurgical evaluations, including EEG studies, are
essential to determine the location of the epileptogenic cortical tissue in patients with partial epilepsy
and neuroimaging alterations (Engel, 1987a). The
instrumentation necessary for the computer-assisted stereotactic resection is expensive and requires trained personnel (Kelly, 1988). Conventional microneurosurgicaltechniques are not readily
extrapolated for this technique (Kelly, 1988). Finally, not all patients appear to experience a worthwhile reduction in seizures after stereotactic lesionectomy (Cascino et al., 1989, 1990). A second
procedure involving resection of the epileptogenic
area may be necessary in these patients to produce
a beneficial effect on seizure outcome andor a reduction in the need for antiepileptic drugs.
Computer-assisted volumetric stereotactic lesionectomy allows pathologic determination of lowgrade tumors and other intra-axial brain-mass lesions and may reduce seizure activity in some patients with intractable partial seizure disorders. This
procedure can be performed with a low associated
morbidity, even for lesions that are relatively inaccessible to standard operative procedures.
Acknowledgment: The author acknowledges the guidance and support of Drs. Frank Sharbrough and Patrick
Kelly, and the secretarial assistance of Ms. Jeanette Connaughty.

REFERENCES
Apuzzo M U , Sabshin JK. Computer tomographic guidance stereotaxis in the management of intracranial mass lesions. Neurosurgery 1983;12:277-85.
Babb TL, Brown WJ. Pathological findings in epilepsy. In: Engel
J Jr, ed. Surgical treatment of the epilepsies. New York:
Raven Press, 1987:511-40.
Epilepsia. Vol. 3I. Suppl. 3. 19W

S44

G . D . CASCINO

Blume WT, Girvin JP, Kaufmann JCE. Childhood brain tumors
presenting as chronic uncontrolled focal seizure disorders.
Ann Neurol 1982;12:538-41.
Cascino GD, Kelly PJ, Hirschorn KA, et al. Stereotactic resection of intra-axial cerebral lesions in partial epilepsy. Mayo
CIin Proc 1990;65:1053-60.
Cascino GD, Kelly PJ, Sharbrough FW,et al. Computer-assisted
volumetric stereotactic resection (lesionectomy) in partial
epilepsy. Epilepsia 1989;30:674.
Dam AM, Fuglsang-Frederiksen A, Svarre-Olsen U, et al. Lateonset epilepsy: etiologies, types of seizure, and value of clinical investigation, EEG, and computerized tomography scan.
Epilepsia 1985;26:227-31.
Daumas-Duport C, Scheithauer BW, Chodkiewicz JP, et al.
Dysembryoplastic neuroepithelial tumor: a surgically curable
tumor of young patients with intractable partial seizures. Neurosurgery 1988;23:545-56.
Drake J, Hoffman HJ,Kobayashi J, et al. Surgical management
of children with temporal lobe epilepsy and mass lesions.
Neurosurgery 1987;21:792-7.
Echlin FA. The supersensitivity of chronically “isolated” cerebral cortex as a mechanism in focal epilepsy. Electroencephalogr Clin Neurophysiol 1959;11:697-722.
Engel J Jr. Approaches to localization of the epileptogenic lesion.
In: Engel J Jr, ed. Surgical treatment of the epilepsies. New
York: Raven Press, 1987a:75-95.
Engel J Jr. Outcome with respect to epileptic seizures. In: Engel
J Jr, ed. Surgical treatment of the epilepsies. New York:
Raven Press, 1987b:553-71.
Engel J Jr. Basic mechanisms of epilepsy. In: Engel J Jr, ed.
Seizures and epilepsy. Philadelphia: F. A. Davis, 1989:71111.
Gloor P. Commentary: approaches to localization of epileptogenic lesion. In: Engel J Jr, ed. Surgical treatment of the
epilepsies. New York: Raven Press, 1987:97-100.
Goldring S, Rich K, Picker S. Experience with gliomas in patients
presenting with a chronic seizure disorder. CIin Neurosurg
1986;33:15-42.
Goldring S,Gregorie EM. Surgical management of epilepsy using
epidural recordings to localize the seizure focus. J Neurosurg
1984;60:457-66.
Hauser WA, Kurland LT. The epidemiology of epilepsy in Rochester, Minnesota, 1935 through 1967. Epilepsia 1975;16:1-66.
Hirsch JF, Sainte Rose C, Pierre-Kahn A, et al. Benign astrocytic
and oligodendrocytic tumors of the cerebral hemispheres in
children. Neurosurgery 1989;70:568-72.
Jackson JH. Localized convulsions from tumor of the brain.
Brain 1882;5:364-74.
Jackson, JH. Case of tumor of the right temporo-sphenoidal lobe,
bearing on the localization of the sense of smell and the interpretation of a particular variety of epilepsy. In: Taylor J,
ed. Selected writings of J . Hughlings Jackson. New York:
Basic Books, 1958:406-11.
Kelly PJ. Computer-assisted stereotaxis: new approaches for the
management of intracranial intra-axial tumors. Neurology
1986;36:535-41.
Kelly PJ, Daumas-Duport C, Kispert DB, et al. Imaging-based

Epilrpsia. Vol. 31, Suppl. 3, 1990

stereotactic serial biopsies in untreated intracranial giial neoplasms. J Neurosurg 1987;66:865-74.
Kelly PJ. Volumetric stereotactic surgical resection of intra-axial
brain mass lesions. Mayo Clin Proc 1988;63: 1186-98.
Ketz E. Brain tumors and epilepsy. In: Vinken PJ, Bruyn GW,
eds. Handbook of clinical neurology, Vol 16. Amsterdam:
North-Holland, 1974:254-69.
Kornblith PL, Walker MD, Cassady JP. Pathology of central nervous system tumors. In: Kornblith PL, Walker MD, Cassady
JR, eds. Neurologic oncology. Philadelphia: J. B. Lippincott,
1987:35-48.
Le Blanc FE, Rasmussen T. Cerebral seizures and brain tumors.
In: Vinken PJ, Bruyn GW, eds. Handbook of clinical neurology, Vol 15. Amsterdam: North-Holland, 1974:295-301.
Lesser R, Liiders H, Dinner D, et al. The location of speech and
writing functions in the frontal language area: results of extraoperative cortical stimulation. Brain 1984;107:275-91.
Lund M. Epilepsy in association with intracranial tumor. A c t a
Neurol Psychiatr Scand 1952;81(suppl):87- 106.
Ojemann G. Brain organization for language from the perspective
of electrical stimulation mapping. Behav Brain Sci 1983;
61: 189-230.
Ojemann G . Surgical treatment of epilepsy. In: Wilkins R, Rengachary S , eds. Neurosurgery. New York: McGraw-Hill,
1985:25 17-27.
Ojemann G, Dodrill C. Verbal memory defects after left temporal
lobectomy for epilepsy. J Neurosurg 1985;62:101-7.
Penfield W, Erickson TC, Tarlov I. Relation of intracranial tumors and symptomatic epilepsy. Arch Neurol Psychiatry
1940~44:300-15.
Penfield W, Jasper H, eds. Epilepsy and the functional anatomy
of the human brain. Boston: Little, Brown andCo., 1954:774.
Rasmussen T. Surgery of epilepsy associated with brain tumors.
In: Purpura DP, Penry JK, Walter RD, eds. Neurosurgical
management of the epilepsies. New York: Raven Press,
1975:227-39. (Advances in neurology, Vol 8).
Rasmussen T, Blundell J. Epilepsy and brain tumor. Clin Neurosurg 1959;7: 138-58.
Shapiro WR, Shapiro JP. Primary brain tumors. In: Asbury AK,
McKhann GM, McDonald WI, eds. Diseases of the nervous
system. Clinical neurobiology. Philadelphia: W. B. Saunders,
1986:1136-54.
Shorvon S. Imaging in the investigation of epilepsy. In: Hopkins
A, ed. Epilepsy. New York: Demos, 1987:201-28.
Spencer DD, Spencer SS, Mattson RH, et al. Intracerebral
masses in patients with intractable partial epilepsy. Neurology 1984;34:432-6.
Sperling MR, Sutherling WW, Nuwer MR. New techniques for
evaluating patients for epilepsy surgery. In: Engel J Jr, ed.
Surgical treatment of the epilepsies. New Y o r k Raven Press,
1987:235-57.
Suarez JC, Sfaello ZM, Guerrero A, et al. Epilepsy and brain
tumors in infancy and adolescence. Child Nerv Sysr
1986;2:169-74.
Wilden JN, Kelly PJ. CT computerized stereotactic biopsy for
low density CT lesions presenting with epilepsy. J Neurol
Neurosurg Psychiatry 1987;50:1302-5.