Epilepsy Research (2014) 108, 701708

journal homepage: www.elsevier.com/locate/epilepsyres

Beyond the lesion: The epileptogenic

networks around cavernous angiomas
Amandine Sevy a, Martine Gavaret a,b,c, Agns Trebuchon a,b,c,
Lisa Vaugier a,b,c, Fabrice Wendling e, Romain Carron d,
Jean Regis d, Patrick Chauvel a,b,c, Aileen Mc Gonigal a,b,c,
Fabrice Bartolomei a,b,c,
a

CHU Timone, Service de Neurophysiologie Clinique, Assistance Publique des Hpitaux de Marseille,
Marseille F-13005, France
b
INSERM, U1106, Marseille F-13005, France
c
Aix Marseille Universit, Facult de Mdecine, Marseille F-13005, France
d
Service de Neurochirurgie fonctionnelle et Strotaxie, Assistance Publique des Hpitaux de Marseille,
Marseille F-13005, France
e
INSERM, U1099, Universit Rennes 1, Rennes, France
Received 17 August 2013; received in revised form 29 January 2014; accepted 28 February 2014
Available online 12 March 2014

KEYWORDS
Cavernous angioma;
Partial epilepsy;
SEEG;
Epileptogenicity
index

Summary
The relationship between epileptogenic lesions and the extension of epileptogenicity is a major challenge in presurgical evaluation of drug resistant epilepsies. In this study,
we aimed at quantifying the epileptogenic properties of brain structures explored by depth
electrodes in patients investigated by stereoelectroencephalography (SEEG) and suffering from
focal drug-resistant epilepsy associated with cavernous angioma (CA). Epileptogenicity of the
perilesional region and distant brain areas was calculated according to the epileptogenicity
index (EI), a technique that allows mathematical quantication of rapid discharges at seizure
onset taking into account the time at which the discharge occurs. Thirteen seizures from 6
patients were studied.

Abbreviations: Am, amygdala; Hip, hippocampus; iTp, internal temporopolar cortex; eTP, external temporopolar cortex; EC, entorhinal
cortex; PRh, perirhinal cortex; TBa, anterior temporobasal cortex; TBp, posterior temporobasal cortex; MTGa, anterior part of middle
temporal gyrus; MTGp, posterior part of middle temporal gyrus; PHC, parahippocampal cortex; STG, superior temporal gyrus; SMA, supplementary motor area; BA6, premotor lateral cortex; OPF, frontal operculum; OFC, orbitofrontal cortex; PFC(F3), prefrontal cortex, inferior
frontal gyrus; PFC (F2), prefrontal cortex, middle frontal gyrus; GC32, cingulate gyrus area 32; CG24, cingulate gyrus area 24; Rol, rolandic
cortex superior(s) or inferior (i) parts; FUS, fusiform gyrus; OC, occipital region; Ins, insular cortex; IPL, inferior parietal lobule; pOP, parietal
operculum.
Corresponding author at: Service de Neurophysiologie Clinique, CHU Timone-264 Rue st Pierre, 13005 Marseille, France.
Tel.: +33 491385833; fax: +33 491385826.
E-mail address: fabrice.bartolomei@ap-hm.fr (F. Bartolomei).
http://dx.doi.org/10.1016/j.eplepsyres.2014.02.018
0920-1211/ 2014 Elsevier B.V. All rights reserved.

702

A. Sevy et al.
Localization of the cavernoma was the frontal lobe (two cases), the temporal lobe (three cases)
or the anterior insula (one case). Visual inspection of the ictal discharge showed that in the
majority of cases (5/6) the perilesional region was either not involved or involved with other
distant sites. Using EI quantication, complex patterns of epileptogenicity were observed in
ve patients. A large number of brain regions out of the lesional region disclosed higher values
than the lesion site. Mean values in the perilesional region and in the extralesional sites were
not signicantly different (p = 0.34). Complex organization of the epileptogenic zone may be
found in drug-resistant CA associated epilepsy. Thus, this result should be borne in mind when
patients with CA and drug resistant epilepsy are investigated. If there is a suspicion of a larger
epileptogenic zone than the lesion, intra-cerebral exploration by SEEG may be required before
surgery that may be guided by the denition of the EZ.
2014 Elsevier B.V. All rights reserved.

Introduction
Cavernous angiomas (CA) are well-circumscribed, mulberrylike vascular malformations that may be found in the central
nervous system in up to 0.5% of the population (Del Curling
et al., 1991). CA can be sporadic or inherited. The common
symptoms are epilepsy, hemorrhages, focal neurological
decits, and headaches. They represent about 520% of all
vascular malformations in the central nervous system. Traditionally seizures have been estimated to occur in 4070%
of patients with CA (Chang et al., 2009). However a recent
report of a large cohort with long-term follow up (2035
patient-years) showed a rather lower prevalence of seizures
at around 30%; these most often occurred without evidence of associated acute hemorrhage (Flemming et al.,
2012). Most seizures are focal, without or with secondary
generalization and are resistant to antiepileptic drugs in
approximately 40% of cases (Chang et al., 2009; Englot et al.,
2011; Moran et al., 1999; Ryvlin et al., 1995).
A number of different mechanisms have been proposed
to cause epilepsy in patients with cavernomas: mass effect,
gliosis, or hemosiderin deposition in the surrounding brain
tissue (Chang et al., 2009). These cases with established
epilepsy should be distinguished from patients who present
only isolated, non-recurring seizures. Surgical resection of
the cavernoma and surrounding tissue is often proposed but
the surgical strategies in patients with CA are still unclear
(Fernandez et al., 2012; von der Brelie and Schramm, 2011).
A large proportion of surgical reports deal with heterogeneous case series, which include patients with various
epileptological contexts, ranging from sporadic episodic
seizures to drug resistant epilepsy (Englot et al., 2011; von
der Brelie and Schramm, 2011). Since the etiology and prognosis are likely to be quite different for these two ends of the
spectrum, global evaluation of outcome is rather difcult.
The rate of success is therefore variable in the literature as recent review has well pointed out (Englot et al.,
2011). In most of the cases, surgical strategy is decided
on neuroimaging data sometimes completed by non-invasive
video EEG recordings. The causes of the failure (occurring
in 3070% of cases) are not well explained. Removal of
the hemosiderin ring surrounding the lesion may improve
outcome but remains debated (Hugelshofer et al., 2011).
Use of elecrocorticography to dene seizure onset zone,
with resection extending beyond the cavernoma and its
hemosiderin ring if necessary, has been associated with

better outcome (Van Gompel et al., 2009). A shorter preoperative history of epilepsy (1 year) was also found to
be an important factor of therapeutic response in recent
meta-analysis (Englot et al., 2011).
Extension of the epileptogenic zone out of the lesional
site could be another explanation for poorer outcome
that has been however rarely addressed. The fact that
widespread networks may be observed in cases of focal
epileptogenic lesion has been demonstrated in focal cortical
dysplasia or dysembryoplastic tumors (Aubert et al., 2009;
Chassoux et al., 2012a, 2012b) and is a probable important
prognostic factor after epilepsy surgery.
To the best of our knowledge, there is no detailed intracerebral EEG study of the epileptogenic zone (EZ) in CA.
In this study, we thus aimed at quantifying the
epileptogenic properties of brain structures explored by
depth electrodes in patients investigated by stereoelectroencephalography (SEEG) and suffering from focal
drug-resistant epilepsy associated with CA. This quantication was performed using the epileptogenicity index (EI).
We determined EI from signals recorded in distinct brain
structures including the perilesional region.

Patients and methods

Patients and SEEG recordings
Patients suffering from drug-resistant focal epilepsies
associated with cavernomas and who had undergone intracerebral recording (stereoelectroencephalography, SEEG)
were included in the study.
They were selected from a retrospective analysis of 295
patients in our epilepsy center database and investigated
with SEEG during presurgical evaluation for drug resistant
epilepsy between 2000 and 2012. Six patients required
invasive recordings after the non-invasive phase for their
epilepsy associated with CA. SEEG was indicated because
the epileptogenic zone was suspected to be larger/or discordant regarding the lesion localization or because of the
localization in or close to eloquent cortex.
SEEG recordings were done according to previous reports
(McGonigal et al., 2008; Talairach et al., 1992). The placement of electrodes was based upon available non-invasive
information providing hypotheses about the localization of
the epileptogenic zone. Therefore, the number of electrodes and their location were dened for each individual

Main clinical data of included patients.

Patients

Age at onset
(years)

Age at SEEG
(years)

Ictal semiology

Cavernoma
localization

Side of
epilepsy

24

33

Fear, epigastric
sensation
LOC, automatisms
Post-ictal aphasia

Left
orbitofrontal

Left

9 (left)

21

Left frontal
(inferior
frontal
gyrus)

Left

7 (left)

28

34

Right dysesthesia
LOC
Left version and
bilateral tonic
posture
Epigastric sensation
Dj vu
LOC
Automatisms

Right
temporal
(neocortex)

Right

7 (right)

36

48

Right
anterior
insula

Right

9 (Right)

14

37

Epigastric sensation
Complex motor
automatisms
Vocalization
Thoracic oppression
Automatisms
LOC

Right
fusiform
gyrus

Right

7 (right)

26

32

Left
temporal
lateral

Left

Hot sensation
LOC, automatisms

Number of
SEEG
electrodes

7(left)

SEEG ndings at seizure

onset

Surgery (FU,
outcome)

LVRD affecting the

orbito-frontal cortex
(cavernoma region) and the
anteromesial temporal
regions. One seizure started
in the perilesional
orbitofrontal cortex
LVRD affecting the premotor
lateral cortex and the
opercular cortex (site of the
cavernoma)

Yes (frontotemporal
cortectomy,1,
IA)

LVRD starting in the

lateral-anterior temporal
cortex (cavernoma region)
and in the entorhinal cortex
and the amygdala
LVRD affecting the temporal
pole and the orbito-frontal
cortex

Nob

Localized beta activity in

the perilesional cortex
(right fusiform and
parahippocampal gyrus)

Yes
(extended
lesionectomy, 2,
IA)
NOc

One seizure started in the

perilesional cortex (left
temporal cortex). Other
seizures started by a
preictal spiking followed by
a LVRD in the hippocampus
and amygdala

Noa

Nob

Beyond the lesion: The epileptogenic networks around cavernous angiomas

Table 1

Abbreviations: LOC: loss of consciousness; LVRD: low voltage rapid discharges.

a Surgery contraindicated.
b Patient declined surgery.
c Awaiting surgical decision.

703

704

A. Sevy et al.

Figure 1 Schematic diagram of SEEG electrodes placement on a lateral view the brain for the six patients. For each patient, MRI
showing the cavernoma (surrounded by green dotted line) and reconstruction of the electrode trajectory from post-implantation
MRI or CT-scan/MRI fusion is indicated. (For interpretation of the references to colour in this gure legend, the reader is referred
to the web version of this article.)

case (Table 1). In each cases however, there was at least

one electrode placed in the perilesional region (Fig. 1).
Finally, MRI was performed to accurately check the anatomical location of each contact along the electrode trajectory.
Intracerebral EEG signals were recorded on a 128 channels
DeltamedTM/Natus system. They were sampled at 512 Hz
and recorded on a hard disk (16 bits/sample) using no digital
lter.

Signal analysis using the epileptogenicity index (EI)

Recently, we proposed a method for quantifying the epileptogenicity of brain structures in patients explored with
depth electrodes and SEEG (Bartolomei et al., 2008, 2010). It
is based on the determination of the epileptogenicity index
(EI) that combines both spectral and temporal parameters
related to the propensity of a brain area to generate rapid

discharges (fast oscillations) and the time for this area to

become involved in the seizure process. Indeed, it is generally accepted that high frequency oscillations and delay
of involvement are relevant indicators of the epileptogenic
nature of the recorded underlying neuronal systems.
The EI is a normalized quantity, ranging from 0 to 1, computed from SEEG recorded signals (see (Bartolomei et al.,
2008) for a more detailed description). A two steps procedure is used for its estimation: rst, over a sliding window,
the signal energy ratio (ER) is computed, between high (
(12.424 Hz) and  (2490 Hz)) and low ( (3.47.4 Hz) and
(7.412.4 Hz)) frequency bands of the EEG; second, we
detect change-points in the ER[n] quantity which is sensitive to frequency changes in the signal. In particular, we
use an optimal algorithm (cumulative sum algorithm or
CUSUM) to automatically determine the time instant
when ER[n] increases, i.e. when  activity (that is predominant in background SEEG signals) changes into 

Beyond the lesion: The epileptogenic networks around cavernous angiomas

705

Figure 2 Figure illustrates the estimation of EI in patient P4. (a) Schematic diagram of SEEG electrodes placement on a lateral
view the brain A: electrode exploring the amygdala (medial leads) and the anterior part of the middle temporal gyrus (MTG) (lateral
leads) B: electrode exploring the anterior hippocampus (medial leads) and the mid part of MTG (lateral leads); TP: electrode
exploring the temporal pole; H: electrode exploring the insula (medial leads) and the anterior part of the superior temporal gyrus
(STG) (lateral leads) OF: electrode exploring the frontal operculum(external leads) and the anterior insula (internal leads); OP:
electrode exploring the parietal operculum(external leads) and the posterior insula (internal leads); Cc: electrode exploring the
frontal premotor cortex (external leads) and the cingulate gyrus (internal leads), Cr: electrode exploring the prefrontal cortex, OR:
electrode exploring the prefrontal cortex (superior leads) and the orbitofrontal cortex (inferior leads). Dashed circle indicates the
location of the cavernoma in the anterior insular cortex (close to the internal leads of the electrode OF). (b) SEEG signal at the
onset of a seizure involving different selected bipolar traces. (c) Corresponding increase in energy ratio (ER[n] which represent the
ratio between high frequencies and low frequencies) is shown in the selected SEEG signals. Each channel discloses variations of the
ER[n]. The Page-Hinkley algorithm provides detection times (vertical red bars) for each channel if involved in the generation of a
rapid discharge. (d) EI values quoted on web chart demonstrating maximal epileptogenicity in the temporal lobe.

activity (that is predominant in SEEG signals during rapid

discharges).
These two steps allow for the computation of the epileptogenicity index EIi from the SEEG signal si recorded from
brain structure Si :
EIi =

N
di +H
1
ER[n],
Ndi N0 + 

>0

n=Nd

where N0 is the time instant corresponding to seizure onset

(dened hereafter), Ndi is the detection time in signal si
recorded from structure Si and H is the duration over which
ER[n] is integrated. Parameter  accounts for the particular
where Si is the rst structure that generates the fast activity
(Ndi = N0 , seizure onset) and avoids division by zero. It was
arbitrarily set to 1. Parameter H was set to be equal to 5 s
as previously dened to detect the onset of rapid discharges
(Bartolomei et al., 2008). In practice, we use a semiautomatic approach: using a handy graphical user interface,
the user can easily inspect and validate automatically

detected change points indicating the accurate onset of

rapid discharges. From this validation performed on a
structure-by-structure basis, the EI is then computed.
Finally, in order to obtain a normalized value ranging
from 0 (no epileptogenicity) to 1 (maximal epileptogenicity) for considered structures Si , EIi values were divided by
the maximal value obtained in each patient. In the sequel,
normalized EIi values are simply denoted by EI values (see
Figs. 2 and 4).
For all selected patients EI was calculated from bipolar
signals recorded from different explored brain regions. We
have selected contacts localized in the gray matter and representative of the different anatomical structures sampled
in each patient (Fig. 4).

Results
Clinical and SEEG data
Six patients with drug resistant focal epilepsies (four men,
two women) associated with cavernoma were included in

706

A. Sevy et al.

Figure 3 SEEG recordings in two patients. (a and b) Two seizures in P1. (a) The seizure starts from TP and OFC and secondarily
involves Hip and Am. (b) The seizure starts in Am and Hip. The OFC channel records the orbito frontal cortex, close to the pericavernoma region. (c and d) Two seizures in P6. (c) The seizure starts from Fus, corresponding to the fusiform gyrus close to the CA.
(d) The seizure starts from the mesial temporal lobe (Tp, EC, Am, Hip) and only secondarily affects the fusiform gyrus.

Figure 4 Epileptogenicity proles in the six patients. EI values, ranging from 0 (no epileptogenicity) to 1 (maximal epileptogenicity)
are given in each studied brain region in each patient for all the studied seizures. The region where the cavernoma (CA) is located
is highlighted in pink. A good concordance between CA region and epileptogenicity prole is seen only in case 5. (For interpretation
of the references to colour in this gure legend, the reader is referred to the web version of this article.)

the study. Localization of the cavernoma was the frontal

lobe (two cases), the temporal lobe (three cases) or the
anterior insula (one case). Mean age at presurgical evaluation was 34.6 years, range 2148 (Table 1). They had
partial seizures for several years (mean (mean 12, range
623)) before entering into presurgical evaluation. Three
patients had secondary generalized tonicclonic seizures.
SEEG recordings with an average 7.8 electrodes allowed to
localize the ictal onset zone in all the patients.

Visual inspection of the ictal discharge showed that in the

majority of cases (5/6) the perilesional region was either
not involved (P4) or involved with other distant sites. Only
one case showed a strict colocalisation of the EZ within
the perilesional region (P5). For two patients, there were
two different types of seizure (P1, P6), with some seizures
in which the lesional site was rst involved while in other
seizures a larger set of structures out of the CA region was
involved (Fig. 3).

Beyond the lesion: The epileptogenic networks around cavernous angiomas

707

Quantication of SEEG signals

Epileptogenicity proles
In order to better analyze and quantify the involvement
of lesional region and other brain regions, we used the EI
method in 13 seizures in the six patients. They represented
all the spontaneous seizures recorded during SEEG monitoring.
In Fig. 4 are indicated for each patient the prole
of EI values (mean of 13 seizures). As suspected from
visual analysis, complex patterns of epileptogenicity were
observed in ve patients.
Patient P1 had middle range value in the frontobasal
region (site of the CA) while maximal epileptogenicity was
observed in the mesial temporal region. Patient P2 had high
value in the perilesional site but other regions of the frontal
lobe depicted high EI values. Patient P3 had relative low
values in the lesion site (lateral temporal cortex) and maximal EI values in the mesial temporal lobe. Patient P4 had
very low value (close to 0) in the perilesional site (anterior
insula) while maximal values were obtained in the temporal
lobe regions. In contrast patient P5 had the highest values
in the perilesional region. Patient P6 had an average prole
of epileptogenicity depicting middle range values into the
perilesional site and maximal within some medial temporal
regions (entorhinal cortex and internal temporal pole).
Intrinsic lesional epileptogenicity versus extralesional
epileptogenicity
We compared the mean EI values obtained in the perilesional
region and in distant regions. Fig. 5a shows the distribution
of values. A large number of brain regions out of the lesional
region disclose higher values than the lesion site. Mean values in the perilesional region and in the extralesional sites
are not signicantly different (p = 0.34, MW).
Fig. 5b illustrates the number of brain regions exhibiting
EI values above 0.4 (NEI 0.4). This threshold has been chosen arbitrarily according to previous studies (Aubert et al.,
2009). This value (NEI 0.4) gives an indication of the extent
of the epileptogenic network in a given patient. In the six
patients, this number ranges from 2 (P5) to 6 (P2), with a
median at 3, indicating that different regions may be epileptogenic in these cases, not limited to the perilesional site.

Discussion
The relationship between epileptogenic lesions and the
extension of epileptogenicity is a major challenge in presurgical evaluation of drug resistant epilepsies. In this study
we have investigated the role of the perilesional region and
distant regions in the organization of the seizure onset zone
in patients with a cavernous malformation, presenting drugresistant epilepsy.
The main result is that the EZ was complex in most cases
(5/6) and largely extended beyond the limits of the perilesional sector. We indeed observed that the seizure onset
may be restricted to the perilesional site but this pattern
was found in only one case and was associated with other
modes of seizure onset in two other cases. In the other
cases the perilesional region was either involved in association with other brain sites or apparently not involved

Figure 5 (b) Histograms showing the number of brain regions

disclosing high EI values (NEI 0.4) in each patient. (a) Scatter
graph of EI values in regions outside the cavernoma regions
(ExCav) and in the perilesional area (CAV).

at the beginning of the seizure. Indeed the lack of concordance between lesion and EZ helps to explain why only
two patients in this series nally underwent surgery for their
drug-resistant epilepsy. This aspect is a possible limitation of
our study and reects the selection bias of this study. However in all the cases, we were able to reasonably dene a
seizure onset zone based on classical criteria in SEEG investigations, in particular because electrographic onset clearly
preceded clinical manifestations.
Our objective was also to focus on patients with CA associated with chronic drug resistant epilepsy. Therefore, our
series is highly selective and not typical of the general population of patients with epilepsy and CA.
In addition to visual analysis, we used quantication of
EEG signal, using the EI method, a method already used
in this context in previous studies (Aubert et al., 2009;
Bartolomei et al., 2008, 2010). This quantication revealed
that the median number of highly epileptogenic structures
was three, suggesting that more than one site exhibited high
epileptogenicity. In addition, values in lesional and extralesional sites were not signicantly different. This result
advocates for an epileptogenicity that is not restricted to
the perilesional region in our cases.
To the best of our knowledge, no detailed intracerebral evaluation of CA is available in the literature. Surgical
strategies in CA are controversial, depending probably on
the context in which the patients are seen. That may explain

708
the results of lesionectomy that range from 30% to 100%
seizure-freedom with an average of 75% as reported in
a recent meta-analysis of surgical ablation of the CA in
patients with epilepsy (Englot et al., 2011). This analysis
reveals that seizure freedom is more frequently associated
with a shorter duration of epilepsy and with drug sensitivity before surgery. A recent Swedish study of intractable
epilepsy associated with either DNET, low grade astrocytoma, ganglioma or cavernoma, found that only shorter
epilepsy duration was associated with better outcome, and
not lesion type (Rydenhag et al., 2013).
Our study points out the potential complexity of the relationship between the CA and the epileptogenic zone. This is
a plausible mechanism of failure when a lesionectomy is performed even when extended to the perilesional hemosiderin
ring. However, these results need to be conrmed in a larger
series and confronted to surgical outcome. The mechanisms
of the elaboration of complex epileptogenic networks are
unclear. However, as proposed in other situations, secondary
epileptogenesis could be involved. This process could take
several years in humans, helping to explain the relationship between the extension of the epileptogenicity and the
duration of epilepsy found in some partial lesional epilepsies
(Aubert et al., 2009; Bartolomei et al., 2010) and between
prognosis and epilepsy duration in patients with CA (Englot
et al., 2011).

Conclusion
This study reveals complex organization of the EZ in a majority of patients with drug-resistant epilepsy and cavernoma
explored by depth electrodes. Thus, this result should be
borne in mind when patients with CA and drug resistant
epilepsy are investigated. If there is a suspicion of a larger
epileptogenic zone than the lesion, intra-cerebral exploration by SEEG may be required before surgery that may
be guided by the denition of the EZ.

References
Aubert, S., Wendling, F., Regis, J., McGonigal, A., Figarella-Branger,
D., Peragut, J.C., Girard, N., Chauvel, P., Bartolomei, F., 2009.
Local and remote epileptogenicity in focal cortical dysplasias
and neurodevelopmental tumours. Brain: A Journal of Neurology
132, 30723086.
Bartolomei, F., Chauvel, P., Wendling, F., 2008. Epileptogenicity of
brain structures in human temporal lobe epilepsy: a quantied
study from intracerebral EEG. Brain: A Journal of Neurology 131,
18181830.
Bartolomei, F., Cosandier-Rimele, D., McGonigal, A., Aubert, S.,
Regis, J., Gavaret, M., Wendling, F., Chauvel, P., 2010. From
mesial temporal lobe to temporoperisylvian seizures: a quantied study of temporal lobe seizure networks. Epilepsia 51,
21472158.
Chang, E.F., Gabriel, R.A., Potts, M.B., Garcia, P.A., Barbaro, N.M.,
Lawton, M.T., 2009. Seizure characteristics and control after

A. Sevy et al.
microsurgical resection of supratentorial cerebral cavernous
malformations. Neurosurgery 65, 3137, discussion 3738.
Chassoux, F., Landre, E., Mellerio, C., Turak, B., Mann, M.W.,
Daumas-Duport, C., Chiron, C., Devaux, B., 2012a. Type II focal
cortical dysplasia: electroclinical phenotype and surgical outcome related to imaging. Epilepsia 53, 349358.
Chassoux, F., Rodrigo, S., Mellerio, C., Landre, E., Miquel, C., Turak,
B., Laschet, J., Meder, J.F., Roux, F.X., Daumas-Duport, C.,
Devaux, B., 2012b. Dysembryoplastic neuroepithelial tumors:
an MRI-based scheme for epilepsy surgery. Neurology 79,
16991707.
Del Curling Jr., O., Kelly Jr., D.L., Elster, A.D., Craven, T.E., 1991.
An analysis of the natural history of cavernous angiomas. Journal
of Neurosurgery 75, 702708.
Englot, D.J., Han, S.J., Lawton, M.T., Chang, E.F., 2011. Predictors
of seizure freedom in the surgical treatment of supratentorial cavernous malformations. Journal of Neurosurgery 115,
11691174.
Fernandez, S., Miro, J., Falip, M., Coello, A., Plans, G., Castaner,
S., Acebes, J.J., 2012. Surgical versus conservative treatment in
patients with cerebral cavernomas and non refractory epilepsy.
Seizure: The Journal of the British Epilepsy Association 21,
785788.
Flemming, K.D., Link, M.J., Christianson, T.J., Brown Jr., R.D.,
2012. Prospective hemorrhage risk of intracerebral cavernous
malformations. Neurology 78, 632636.
Hugelshofer, M., Acciarri, N., Sure, U., Georgiadis, D., Baumgartner, R.W., Bertalanffy, H., Siegel, A.M., 2011. Effective surgical
treatment of cerebral cavernous malformations: a multicenter
study of 79 pediatric patients. Journal of Neurosurgery. Pediatrics 8, 522525.
McGonigal, A., Gavaret, M., Da Fonseca, A.T., Guye, M., Scavarda, D., Villeneuve, N., Regis, J., Bartolomei, F., Chauvel, P.,
2008. MRI-negative prefrontal epilepsy due to cortical dysplasia explored by stereoelectroencephalography (SEEG). Epileptic
Disorders: International Epilepsy Journal with Videotape 10,
330338.
Moran, N.F., Fish, D.R., Kitchen, N., Shorvon, S., Kendall, B.E.,
Stevens, J.M., 1999. Supratentorial cavernous haemangiomas
and epilepsy: a review of the literature and case series. Journal
of Neurology, Neurosurgery, and Psychiatry 66, 561568.
Rydenhag, B., Flink, R., Malmgren, K., 2013. Surgical outcomes in
patients with epileptogenic tumours and cavernomas in Sweden:
good seizure control but late referrals. Journal of Neurology,
Neurosurgery, and Psychiatry 84, 4953.
Ryvlin, P., Mauguiere, F., Sindou, M., Froment, J.C., Cinotti,
L., 1995. Interictal cerebral metabolism and epilepsy in cavernous angiomas. Brain: A Journal of Neurology 118 (Pt 3),
677687.
Talairach, J., Bancaud, J., Bonis, A., Szikla, G., Trottier, S., Vignal,
J.P., Chauvel, P., Munari, C., Chodkievicz, J.P., 1992. Surgical therapy for frontal epilepsies. Advances in Neurology 57,
707732.
Van Gompel, J.J., Rubio, J., Cascino, G.D., Worrell, G.A., Meyer,
F.B., 2009. Electrocorticography-guided resection of temporal
cavernoma: is electrocorticography warranted and does it alter
the surgical approach? Journal of Neurosurgery 110, 11791185.
von der Brelie, C., Schramm, J., 2011. Cerebral cavernous malformations and intractable epilepsy: the limited usefulness of
current literature. Acta Neurochirurgica 153, 249259.