J Neurosurg 115:1169–1174, 2011

Predictors of seizure freedom in the surgical treatment of

supratentorial cavernous malformations

Clinical article
Dario J. Englot, M.D., Ph.D., Seunggu J. Han, M.D., Michael T. Lawton, M.D.,
and Edward F. Chang, M.D.

Department of Neurological Surgery, University of California, San Francisco, California

Object. Seizures are the most common presenting symptom of supratentorial cerebral cavernous malformations
(CCMs) and progress to medically refractory epilepsy in 40% of patients. Predictors of seizure freedom in the resec-
tion of CCMs are incompletely understood.
Methods. The authors systematically reviewed the published literature on seizure freedom following the resec-
tion of supratentorial CCMs in patients presenting with seizures. Seizure outcomes were stratified across 12 potential
prognostic variables. A total of 1226 patients with supratentorial CCMs causing seizures were identified across 31
predominantly retrospective studies; 361 patients had medically refractory epilepsy.
Results. Seventy-five percent of the patients were seizure free after microsurgical lesion removal, whereas 25%
continued to have seizures. All patients had had preoperative seizures and > 6 months of postoperative follow-up.
Modifiable predictors of postoperative seizure freedom included gross-total resection (OR 36.6, 95% CI 8.5–157.5)
and surgery within 1 year of symptom onset (OR 1.83, 95% CI 1.30–2.58). Additional prognostic indicators of a
favorable outcome were a CCM size < 1.5 cm (OR 15.4, 95% CI 5.2–45.4), the absence of multiple CCMs (OR 2.02,
95% CI 1.13–3.60), medically controlled seizures (OR 2.38, 95% CI 1.29–4.39), and the lack of secondarily general-
ized seizures (OR 3.33, 95% CI 2.09–5.30). Other factors, including extended resection of the hemosiderin ring, were
not significantly predictive.
Conclusions. In the surgical treatment of supratentorial CCMs, gross-total resection and early operative inter-
vention may improve seizure outcome. While surgery should not be considered the first-line treatment for CCM-
related epilepsy, it is important to understand the variables associated with seizure freedom in CCM resection given
the considerable morbidity and diminished quality of life associated with epilepsy. (DOI: 10.3171/2011.7.JNS11536)

Key Words      •      cavernoma      •      cavernous angioma      •      epilepsy      •

resection      •      seizure      •      vascular disorders

C
erebral cavernous malformations are endothe- important and often underappreciated treatment goal in
lial-lined vascular malformations composed of managing these lesions.
dilated sinusoids filled with blood and no inter- Factors associated with seizure control in the re-
vening brain tissue and may be associated with intracra- section of supratentorial CCMs remain incompletely
nial hemorrhage.4,39,42 Seizures, the most common pre- understood as a result of patient diversity, small sample
senting symptom of supratentorial CCMs, are thought to sizes, and disagreement about operative approaches. For
arise from the excitotoxic effects of blood products on instance, there is controversy regarding whether exci-
perilesional parenchyma.52–54 Overall, epilepsy affects sion of the hemosiderin-stained parenchymal ring typi-
35%–70% of patients with CCMs, with approximately cally surrounding CCMs is necessary to achieve seizure
40% of individuals progressing to medically refractory freedom6,7,28,49,56,57 and whether an expedited surgical ap-
epilepsy.13,37,44 It is well known that epilepsy can dramati- proach predicts improved seizure outcome.1,8,14 Neither is
cally diminish quality of life with serious morbidity and it fully understood how the preoperative seizure profile
that AEDs frequently have unfavorable side-effect pro- might predict postoperative seizure recurrence.13,37 The
files.10,18,46 Therefore, while most authors examining the systematic consolidation and analysis of predictors of sei-
natural history and treatment of CCMs focus on the risk zure freedom as reported across various smaller studies
of intracranial hemorrhage,34,39,41 abolishing seizures is an may help to clarify these issues.
Our goal was to perform a quantitative and compre-
Abbreviations used in this paper: AED = antiepileptic drug; CCM hensive systematic literature review of seizure outcomes
= cerebral cavernous malformation; ECoG = electrocorticography; following the resection of supratentorial CCMs associat-
GTR = gross-total resection. ed with seizures. We analyzed the proportion of patients

J Neurosurg / Volume 115 / December 2011 1169

 D. J. Englot et al.

in whom complete seizure freedom was achieved after Begg funnel plot method was applied to detect the pos-
surgery, and we stratified outcomes according to several sible effect of publication bias. Missing values were ad-
potential prognostic variables. dressed by listwise deletion, and multiple imputation was
not used. Given the number of comparisons being made,
the level of significance was set at 0.02 for all analyses.
Methods Statistical analysis was performed using SPSS version 17
Article Selection and Data Extraction software (SPSS, Inc.).
We conducted a PubMed search of articles published
between 1985 and February 2011 by using the following Results
search terms alone and in combination: “seizure,” “epi-
We identified 1226 patients across 31 studies on sei-
lepsy,” “cavernous malformation,” “cavernous angioma,” zure control following the resection of supratentorial
“cavernous hemangioma,” “cavernoma,” “surgery,” and CCMs.1–3,5,6,8,9,13,15,16,19–21,23,24,26–29,33,35,37,38,43,47,49–51,55–57 Three
“resection.” References from all selected papers were studies identified on initial query were excluded because
further examined for additional suitable studies and to of a sample size smaller than 5 patients, the lack of seizure
identify possible patient duplication. We selected only outcome disaggregation, or data redundancy, respectively.
those reports of patients harboring supratentorial CCMs All included papers were retrospective analyses, except for
causing preoperative seizures, in which postoperative one prospective observational study, and are listed in Table
seizure status was a primary outcome and the follow-up 1. No controlled trials of seizure outcomes in CCM resec-
spanned a minimum of 6 months for all patients. Studies tion were identified. All patients had experienced lesion-
were excluded if they included fewer than 5 patients, re- related seizures preoperatively and were followed up for at
ported only redundant data, did not disaggregate seizure least 6 months after surgery, with a mean follow-up rang-
outcomes, were not written in English, or consisted of a ing from > 12 to 97 months within the individual studies.
review. Studies that considered lesions other than CCMs Across individual data sets, 33%–100% of patients were
or examined infratentorial lesions were included in part seizure free after resection, as summarized in Fig. 1. Over-
only if the outcomes related to other lesion types could all, 915 patients (75%) were free of seizures after surgery
be excluded. Outcomes from all studies were disaggre- (Engel Class I), whereas 311 (25%) continued to have sei-
gated across 12 variables where possible: 1) patient age zures (Engel Classes II–IV).
at surgery (< 18 vs ≥ 18 years old); 2) sex; 3) CCM loca- Seizure outcomes were stratified across 12 variables
tion (temporal lobe vs extratemporal); 4) CCM laterality; of interest where disaggregation was possible, as summa-
5) CCM size (maximum diameter < 1.5 vs ≥ 1.5 cm); 6) rized in Table 2. We first looked at patient demographics
CCM multiplicity (resection of solitary vs multiple le- and observed similar rates of postoperative seizure free-
sions); 7) preoperative seizure control on medication dom among pediatric versus adult patients (75% vs 73%
(controlled vs refractory); 8) seizure semiology (partial seizure free, respectively) and among female versus male
seizures only vs secondarily generalized seizures); 9) du- patients (71% each). Next, we examined factors related to
ration of epilepsy before surgery (≤ 1 vs > 1 year); 10) the lesion. There was no statistical difference in seizure
extent of resection (gross total vs subtotal); 11) whether outcomes between patients with temporal lobe versus ex-
perilesional hemosiderin-stained parenchyma was ex- tratemporal CCMs, nor with left versus right hemispheric
cised (lesionectomy only vs additional corticectomy); and lesions. However, we did find that individuals with smaller
12) whether awake ECoG was used intraoperatively. Ce- (< 1.5 cm) CCMs attained a dramatically higher seizure
rebral cavernous malformation size was dichotomized at control rate (86%) than did those with a lesion size ≥ 1.5
1.5 cm because prior evidence has suggested a more fa- cm (29% seizure free; p < 0.001). Resection of a solitary
vorable postoperative seizure outcome with lesions below CCM was also predictive of a favorable outcome: while
this size.12 Gross-total resection was defined based on the 71% of those with a solitary lesion became seizure free
lack of residual lesion both intraoperatively and on post- after surgery, only 55% of those with multiple resected
operative imaging. Seizure outcomes were defined using CCMs attained this outcome (p < 0.02).
the Engel classification scheme.22 We next analyzed whether characteristics related
Statistical Analysis to epilepsy predicted the postoperative seizure status.
It was noted that individuals with preoperative seizures
Separate meta-analyses were performed on calculat- controlled with AEDs attained a higher rate of seizure
ed pooled rates of seizure freedom (Engel Class I) versus freedom after CCM resection (84%) than did those with
persistent seizures (Engel Class II–IV) for each variable. medically refractory epilepsy (69%; p < 0.01; Table 2).
Initial between-group comparisons were performed us- Seizure semiology also predicted variable outcomes, as
ing the Pearson chi-square test. Binary logistic regression 81% of patients experiencing only partial seizures be-
analysis was then used to evaluate the variables found to came seizure free postoperatively, compared with only
be significant on chi-square testing by using a backward 56% of individuals with secondarily generalized events
stepwise approach. Odds ratios were calculated with a (p < 0.001). Furthermore, a shorter preoperative history
95% confidence interval. The Friedman test and Kendall of seizures (≤ 1 year) prognosticated a better result (81%
W test were used to analyze data heterogeneity between seizure freedom) than did a longer history (> 1 year) of
studies to ensure that a fixed-effects model was appropri- epilepsy (p < 0.001).
ate, and the Cox and Snell R-squared values were calcu- Finally, factors related to surgical intervention were
lated to ensure goodness-of-fit of the applied model. The examined for possible outcome prognostication. Gross-

1170 J Neurosurg / Volume 115 / December 2011

Seizures after cavernoma resection
TABLE 1: Studies included in analysis of seizure freedom
following treatment for supratentorial CCMs

Authors & Year No. of Cases Mean Follow-Up (mos)

Acciarri et al., 1995  36 71
Acciarri et al., 2009  29 57
Alexiou et al., 2009  14 71
Baumann et al., 2006  31 34
Baumann et al., 2007 168 25
Cappabianca et al., 1997  35 >24
Casazza et al., 1996  47 48
Chang et al., 2009  44 32†
Cohen et al., 1995  50 60
Consales et al., 2010*   5 48†
D’Angelo et al., 2006  69 55 Fig. 1.  Scatterplot demonstrating seizure freedom after CCM resec-
Daglioglu et al., 2010   6 26 tion across 31 studies. Each point on the plot represents 1 study, orga-
nized from 1 to 31 in order of decreasing sample size. Seizure freedom
Dodick et al., 1994  20 36 (Engel Class I outcome) ranged from 33% to 100%. The summary box-
Ferrier et al., 2007  19 >12 plot (right) depicts the median (center line), 25th percentile, and 99th
Ferroli et al., 2006 163 48 percentile of reported seizure freedom rates across all studies, with
outliers (X).
Folkersma & Mooij, 2001   7 40
Giulioni et al., 1995  11 68
of 15.4 (95% CI 5.2–45.4) and 2.02 (95% CI 1.13–3.60),
Hammen et al., 2007  30 24 respectively. Positive predictors related to preoperative
Hsu et al., 2007  15 92 seizure characteristics included adequate medical control
Kivelev et al., 2011  39 72† on AEDs (OR 2.38, 95% CI 1.29–4.39), the presence of
Kraemer et al., 1998  15 66 partial seizures only (OR 3.33, 95% CI 2.09–5.30), and
Moran et al., 1999  16 38 symptom duration ≤ 1 year (OR 1.83, 95% CI 1.30–2.58).
Finally, achieving a gross-total lesionectomy also prog-
Noto et al., 2005  15 64
nosticated seizure control (OR 36.6, 95% CI 8.5–157.5).
Rocamora et al., 2009  11 >24
Shih & Pan, 2005  14 46
Stavrou et al., 2008  57 97 Discussion
Upchurch et al., 2010  12 51
Seizures are the most common presenting symptom
Van Gompel et al., 2009  87 37 in patients with supratentorial CCMs and can progress to
Xia et al., 2009  24 29 medically refractory epilepsy in approximately 40% of
Yeon et al., 2009  60 31 cases.13,37,44 While most studies of CCMs focus on the risk
Zevgaridis et al., 1996  77 39 of intracranial hemorrhage, eliminating the seizure bur-
den in patients with cavernoma and epilepsy can reduce
*  Prospective observational study. Other studies are all retrospective. disability and improve quality of life, and thus is an im-
† Median. portant goal of treatment. To identify the epidemiological
and therapeutic predictors of seizure freedom in the re-
total lesionectomy resulted in a dramatically better section of supratentorial CCMs, we performed a system-
chance of seizure control (77%) than did subtotal CCM atic literature review of 1226 CCM patients with lesion-
resection, after which only 8% of patients were seizure related seizures and stratified the seizure outcomes across
free (p < 0.001; Table 2). However, patients were equally various potentially prognostic variables. Overall, 75% of
likely to stop seizing after undergoing lesionectomy with patients became seizure free after resection, suggesting
or without additional excision of surrounding hemosid- that a microsurgical approach often leads to a favorable
erin fringe (75% vs 76% seizure free, respectively). In outcome but also indicating considerable room for im-
determining whether intraoperative ECoG affected out- provement. Most importantly, we found that potentially
come, we noted that its use led to a somewhat higher rate modifiable predictors of postoperative seizure control in-
of seizure freedom (78%) than did surgery without ECoG cluded both GTR (vs subtotal lesionectomy) and a shorter
(68%), but this difference did not reach our threshold for preoperative history of epilepsy (≤ 1 year). These factors
statistical significance (p = 0.03). reveal that achieving a gross-total lesionectomy and early
Binary logistic regression analyses were used to fur- intervention are important goals in cases in which resec-
ther scrutinize the 6 variables found to be statistically sig- tion is deemed as an appropriate and safe treatment for a
nificant on chi-square testing, as summarized in Table 3. CCM. Gross-total resection can be challenging in deep
A CCM size < 1.5 cm and a solitary lesion significantly or eloquent brain regions, but various microsurgical ap-
predicted postoperative seizure freedom, with odds ratios proaches and trajectories have been recently described

J Neurosurg / Volume 115 / December 2011 1171

 D. J. Englot et al.
TABLE 2: Seizure outcomes stratified across factors of interest TABLE 3: Predictors of seizure freedom after CCM resection

No. (%) Parameter OR 95% CI p Value

Engel Engel Classes CCM size <1.5 cm 15.4 5.2–45.4 <0.001
Parameter Class I II–IV c2 p Value
single CCM only 2.02 1.13–3.60 <0.02
age in yrs seizures medically controlled 2.38 1.29–4.39 <0.01
  <18 71 (75) 24 (25) 0.1 0.79 partial seizures only 3.33 2.09–5.30 <0.001
  ≥18 228 (73) 84 (27) seizures for ≤1 yr 1.83 1.30–2.58 <0.001
sex gross-total lesionectomy 36.6 8.5–157.5 <0.001
 F 98 (71) 41 (29) 0.0 0.90
 M 100 (71) 40 (29)
stained parenchyma surrounding the lesion. Some groups
CCM location have reported improved seizure outcomes with an ex-
 temporal 194 (69) 87 (31) 0.4 0.61 tended lesionectomy including the hemosiderin ring,6,28,49
 extratemporal 274 (71) 111 (29) whereas others have demonstrated no additional benefit
CCM laterality over pure lesionectomy.56,57 Overall, we found no differ-
 lt 63 (63) 37 (37) 2.5 0.11
ence in the rates of seizure freedom after pure lesionec-
tomy versus inclusion of the hemosiderin fringe. This
 rt 65 (74) 23 (26) finding suggests that a more extensive resection beyond
CCM size in cm gross-total lesionectomy may not confer any additional
  <1.5 43 (86) 7 (14) 28.9 <0.001† benefit in seizure control but may be associated with in-
  ≥1.5 10 (29) 25 (71) creased neurological morbidity. Given the conflicting
CCM multiplicity findings in previous reports, however, additional prospec-
tive data should be sought to further clarify this issue.
 single 410 (71) 167 (29) 5.8 <0.02† Not surprisingly, it was also noted that a greater dis-
 multiple 28 (55) 23 (45) ease burden was associated with higher rates of postop-
seizure control* erative seizures. Patients with multiple CCMs were less
 controlled 73 (84) 14 (16) 8.0 <0.01† likely to attain seizure freedom after resection than were
 refractory 248 (69) 113 (31) those who had harbored a single cavernoma. Moreover,
both the presence of secondarily generalized seizures and
seizure semiology
poor seizure control on AEDs predicted a worse outcome.
  partial only 158 (81) 38 (19) 26.8 <0.001† In mesial temporal lobe epilepsy, the most common form
 generalized 95 (56) 76 (44) of epilepsy, secondary generalization has been associated
duration of epilepsy with postoperative seizure recurrence,30,31,36 and the pres-
  ≤1 year 313 (81) 72 (19) 12.3 <0.001† ence of generalized seizures has been linked to more sig-
nificant mesial temporal atrophy.11,17 It is therefore plau-
  >1 year 256 (70) 108 (30)
sible that CCM patients with more severe epileptic events
extent of lesionectomy have higher rates of dual pathology,25,48 which may con-
 GTR 506 (77) 152 (23) 57.3 <0.001† tinue to drive seizures after lesionectomy. More evidence
 subtotal 2 (8) 22 (92) will be needed to further evaluate this possibility.
hemosiderin excision No differences in seizure outcome were observed in
adults versus children, males versus females, or with cav-
  lesionectomy only 224 (76) 71 (24) 0.1 0.82
ernomas located in the left versus the right hemisphere
  w/ corticectomy 352 (75) 116 (25) or the temporal lobe versus the extratemporal region. Fi-
intraop ECoG nally, the use of intraoperative ECoG was associated with
 used 96 (78) 27 (22) 4.6 0.03 a 10% higher rate of postoperative seizure control, but
  not used 265 (68) 125 (32) this difference was not significant. Note, however, that it
total 915 (75) 310 (25)
is not known whether cases in which ECoG is used are
associated a priori with more severe epilepsy or with a
*  Seizures medically controlled or refractory preoperatively. lesion less amenable to GTR given the involvement of
†  Significant value (p < 0.02). eloquent structures, and thus potentially confounding this
result. Furthermore, insufficient data were available to
specifically investigate whether ECoG might be of var-
to facilitate safe access to supratentorial CCMs in these ied benefit in patients with medically refractory versus
locations.12 Finally, our finding that patients with larger controlled epilepsy. Prospective investigations with com-
CCMs (≥ 1.5 cm) continued to seize more often than parable patient characteristics will be needed to further
those with smaller lesions (< 1.5 cm) also supports early address the benefit of intraoperative ECoG for epilepsy
surgical treatment, as some cavernomas have demonstrat- related to CCMs.
ed the potential to grow in size over time.32,40,45 There are several limitations to the present system-
One controversy in the operative treatment of su- atic review. While these data represent a useful summary
pratentorial CCMs is whether to excise the hemosiderin- of seizure control following supratentorial CCM resec-

1172 J Neurosurg / Volume 115 / December 2011

Seizures after cavernoma resection

tion in a large patient population, they are limited by the landri G, et al: Cavernous malformations of the central ner-
quality of the data published by others. We cannot ascer- vous system in the pediatric age group. Pediatr Neurosurg
tain the validity of subjectively defined variables reported 45:81–104, 2009
  2.  Acciarri N, Giulioni M, Padovani R, Galassi E, Gaist G: Sur-
in the original articles, which may have an important in- gical management of cerebral cavernous angiomas causing
fluence on our findings. All except one of the analyzed epilepsy. J Neurosurg Sci 39:13–20, 1995
papers featured retrospective uncontrolled studies, and 3   3.  Alexiou GA, Mpairamidis E, Sfakianos G, Prodromou N: Sur-
articles included fewer than 10 patients, introducing the gical management of brain cavernomas in children. Pediatr
possibility of bias in result reporting and interpretation. Neurosurg 45:375–378, 2009
Moreover, the exclusion of non-English papers may lead   4.  Awad I, Jabbour P: Cerebral cavernous malformations and
to the omission of useful data. As no controlled trials of epilepsy. Neurosurg Focus 21(1):e7, 2006
seizure outcome in CCM resection have been performed   5.  Baumann CR, Acciarri N, Bertalanffy H, Devinsky O, Elger
CE, Lo Russo G, et al: Seizure outcome after resection of su-
to our knowledge, this study cannot be considered a for- pratentorial cavernous malformations: a study of 168 patients.
mal meta-analysis of trial data, but rather is a systematic Epilepsia 48:559–563, 2007
review of available literature. While a single, large pro-   6.  Baumann CR, Schuknecht B, Lo Russo G, Cossu M, Citte-
spective series with post hoc testing of potential interac- rio A, Andermann F, et al: Seizure outcome after resection of
tions would certainly be welcomed, the strength of the cavernous malformations is better when surrounding hemo-
present evaluation lies in the ability to pool a large num- siderin-stained brain also is removed. Epilepsia 47:563–566,
ber of cases, which would be difficult to achieve even in 2006
a multiinstitutional trial. Finally, many interesting issues   7.  Buckingham MJ, Crone KR, Ball WS, Berger TS: Manage-
ment of cerebral cavernous angiomas in children presenting
were not addressed in our study, such as a comparative with seizures. Childs Nerv Syst 5:347–349, 1989
evaluation of microsurgery versus stereotactic radiosur-   8.  Cappabianca P, Alfieri A, Maiuri F, Mariniello G, Cirillo S,
gery to treat CCMs causing seizures, but will certainly de Divitiis E: Supratentorial cavernous malformations and
contribute to excellent questions in future investigations. epilepsy: seizure outcome after lesionectomy on a series of 35
patients. Clin Neurol Neurosurg 99:179–183, 1997
  9.  Casazza M, Broggi G, Franzini A, Avanzini G, Spreafico
Conclusions R, Bracchi M, et al: Supratentorial cavernous angiomas and
epileptic seizures: preoperative course and postoperative out-
In summary, CCMs are vascular brain lesions that come. Neurosurgery 39:26–34, 1996
most commonly appear with seizures and often result in 10.  Cascino GD: When drugs and surgery don’t work. Epilepsia
medically refractory epilepsy. In a systematic review of 49 (Suppl 9):79–84, 2008
1226 patients with supratentorial CCMs, we found that 11.  Cendes F, Andermann F, Gloor P, Gambardella A, Lopes-
early operative intervention and GTR significantly im- Cendes I, Watson C, et al: Relationship between atrophy of
proved seizure outcome. Individuals with a single lesion the amygdala and ictal fear in temporal lobe epilepsy. Brain
and smaller lesions, as well as those with pharmacologi- 117:739–746, 1994
cally controlled epilepsy and the absence of secondarily 12.  Chang EF, Gabriel RA, Potts MB, Berger MS, Lawton MT:
Supratentorial cavernous malformations in eloquent and deep
generalized events, were also more likely to attain post- locations: surgical approaches and outcomes. Clinical article.
operative seizure freedom. Resection should not be con- J Neurosurg 114:814–827, 2011
sidered a first-line treatment for CCM-related epilepsy, 13.  Chang EF, Gabriel RA, Potts MB, Garcia PA, Barbaro NM,
because seizures can often be controlled with AEDs, and Lawton MT: Seizure characteristics and control after micro-
because many patients are not good surgical candidates. surgical resection of supratentorial cerebral cavernous mal-
However, given the considerable effects that epilepsy can formations. Neurosurgery 65:31–38, 2009
have on a patient’s quality of life, it is critical to under- 14.  Churchyard A, Khangure M, Grainger K: Cerebral cavernous
stand the demographic and treatment variables associated angioma: a potentially benign condition? Successful treat-
ment in 16 cases. J Neurol Neurosurg Psychiatry 55:1040–
with seizure control in CCM resection and to incorporate 1045, 1992
that knowledge into our neurosurgical practices and pa- 15.  Cohen DS, Zubay GP, Goodman RR: Seizure outcome after
tient discussions. lesionectomy for cavernous malformations. J Neurosurg 83:
237–242, 1995
Disclosure 16.  Consales A, Piatelli G, Ravegnani M, Pavanello M, Striano
P, Zoli ML, et al: Treatment and outcome of children with
The authors report no conflict of interest concerning the mate- cerebral cavernomas: a survey on 32 patients. Neurol Sci 31:
rials or methods used in this study or the findings specified in this 117–123, 2010
paper. 17.  Cook MJ, Fish DR, Shorvon SD, Straughan K, Stevens JM:
Author contributions to the study and manuscript prepara- Hippocampal volumetric and morphometric studies in frontal
tion include the following. Conception and design: Chang, Englot. and temporal lobe epilepsy. Brain 115:1001–1015, 1992
Acquisition of data: Englot. Analysis and interpretation of data: 18.  Cramer JA, Mintzer S, Wheless J, Mattson RH: Adverse ef-
all authors. Drafting the article: Englot. Critically revising the fects of antiepileptic drugs: a brief overview of important is-
ar­ticle: all authors. Reviewed submitted version of manuscript: all sues. Expert Rev Neurother 10:885–891, 2010
au­thors. Approved the final version of the manuscript on behalf of 19.  D’Angelo VA, De Bonis C, Amoroso R, Cali A, D’Agruma L,
all authors: Chang. Statistical analysis: Englot. Study supervision: Guarnieri V, et al: Supratentorial cerebral cavernous malfor-
Chang, Law­ton. mations: clinical, surgical, and genetic involvement. Neuro-
surg Focus 21(1):e9, 2006
References 20.  Daglioglu E, Ergungor F, Polat E, Nacar O: Microsurgical re-
section of supratentorial cerebral cavernomas. Turk Neuro-
  1.  Acciarri N, Galassi E, Giulioni M, Pozzati E, Grasso V, Pa- surg 20:348–352, 2010

J Neurosurg / Volume 115 / December 2011 1173

 D. J. Englot et al.

21.  Dodick DW, Cascino GD, Meyer FB: Vascular malformations Growth, subsequent bleeding, and de novo appearance of ce-
and intractable epilepsy: outcome after surgical treatment. rebral cavernous angiomas. Neurosurgery 38:662–670, 1996
Mayo Clin Proc 69:741–745, 1994 41.  Robinson JR, Awad IA, Little JR: Natural history of the cav-
22.  Engel J Jr, Van Ness PC, Rasmussen TB, Ojemann LM: Out- ernous angioma. J Neurosurg 75:709–714, 1991
come with respect to epileptic seizures, in Engel J Jr (ed): Sur- 42.  Robinson JR Jr, Awad IA, Masaryk TJ, Estes ML: Pathologi-
gical Treatment of the Epilepsies, ed 2. New York: Raven cal heterogeneity of angiographically occult vascular malfor-
Press, 1993, pp 609–621 mations of the brain. Neurosurgery 33:547–555, 1993
23.  Ferrier CH, Aronica E, Leijten FS, Spliet WG, Boer K, van 43.  Rocamora R, Mader I, Zentner J, Schulze-Bonhage A: Epi-
Rijen PC, et al: Electrocorticography discharge patterns in lepsy surgery in patients with multiple cerebral cavernous
patients with a cavernous hemangioma and pharmacoresistent malformations. Seizure 18:241–245, 2009
epilepsy. J Neurosurg 107:495–503, 2007 44.  Ryvlin P, Mauguière F, Sindou M, Froment JC, Cinotti L: In-
24.  Ferroli P, Casazza M, Marras C, Mendola C, Franzini A, terictal cerebral metabolism and epilepsy in cavernous angio-
Broggi G: Cerebral cavernomas and seizures: a retrospective mas. Brain 118:677–687, 1995
study on 163 patients who underwent pure lesionectomy. Neu- 45.  Scott RM, Barnes P, Kupsky W, Adelman LS: Cavernous an-
rol Sci 26:390–394, 2006 giomas of the central nervous system in children. J Neuro-
25.  Fish DR, Spencer SS: Clinical correlations: MRI and EEG. surg 76:38–46, 1992
Magn Reson Imaging 13:1113–1117, 1995 46.  Sheth RD: Adolescent issues in epilepsy. J Child Neurol 2 (17
26.  Folkersma H, Mooij JJ: Follow-up of 13 patients with surgi- Suppl):2S23–2S27, 2002
cal treatment of cerebral cavernous malformations: effect on 47.  Shih YH, Pan DH: Management of supratentorial cavernous
epilepsy and patient disability. Clin Neurol Neurosurg 103: malformations: craniotomy versus gammaknife radiosurgery.
67–71, 2001 Clin Neurol Neurosurg 107:108–112, 2005
27.  Giulioni M, Acciarri N, Padovani R, Galassi E: Results of sur- 48.  Spencer S, Huh L: Outcomes of epilepsy surgery in adults and
gery in children with cerebral cavernous angiomas causing children. Lancet Neurol 7:525–537, 2008
epilepsy. Br J Neurosurg 9:135–141, 1995 49.  Stavrou I, Baumgartner C, Frischer JM, Trattnig S, Knosp E:
28.  Hammen T, Romstöck J, Dörfler A, Kerling F, Buchfelder M, Long-term seizure control after resection of supratentorial
Stefan H: Prediction of postoperative outcome with special cavernomas: a retrospective single-center study in 53 patients.
respect to removal of hemosiderin fringe: a study in patients Neurosurgery 63:888–897, 2008
with cavernous haemangiomas associated with symptomatic 50.  Upchurch K, Stern JM, Salamon N, Dewar S, Engel J Jr,
epilepsy. Seizure 16:248–253, 2007 Vinters HV, et al: Epileptogenic temporal cavernous malfor-
29.  Hsu PW, Chang CN, Tseng CK, Wei KC, Wang CC, Chuang mations: operative strategies and postoperative seizure out-
CC, et al: Treatment of epileptogenic cavernomas: surgery comes. Seizure 19:120–128, 2010
ver­sus radiosurgery. Cerebrovasc Dis 24:116–121, 2007 51.  Van Gompel JJ, Rubio J, Cascino GD, Worrell GA, Meyer FB:
30.  Janszky J, Janszky I, Schulz R, Hoppe M, Behne F, Pannek Electrocorticography-guided resection of temporal caver-
HW, et al: Temporal lobe epilepsy with hippocampal sclero- noma: is electrocorticography warranted and does it alter the
sis: predictors for long-term surgical outcome. Brain 128: surgical approach? Clinical article. J Neurosurg 110:1179–
395–404, 2005 1185, 2009
31.  Jeong SW, Lee SK, Hong KS, Kim KK, Chung CK, Kim H: 52.  von Essen C, Rydenhag B, Nyström B, Mozzi R, van Gelder N,
Prognostic factors for the surgery for mesial temporal lobe ep- Hamberger A: High levels of glycine and serine as a cause of
ilepsy: longitudinal analysis. Epilepsia 46:1273–1279, 2005 the seizure symptoms of cavernous angiomas? J Neurochem
32.  Jung KH, Chu K, Jeong SW, Park HK, Bae HJ, Yoon BW: 67:260–264, 1996
Cerebral cavernous malformations with dynamic and pro- 53.  Washington CW, McCoy KE, Zipfel GJ: Update on the natu-
gressive course: correlation study with vascular endothelial ral history of cavernous malformations and factors predicting
growth factor. Arch Neurol 60:1613–1618, 2003 aggressive clinical presentation. Neurosurg Focus 29(3):E7,
33.  Kivelev J, Niemelä M, Blomstedt G, Roivainen R, Lehecka 2010
M, Hernesniemi J: Microsurgical treatment of temporal lobe 54.  Williamson A, Patrylo PR, Lee S, Spencer DD: Physiology of
cavernomas. Acta Neurochir (Wien) 153:261–270, 2011 human cortical neurons adjacent to cavernous malformations
34.  Kondziolka D, Lunsford LD, Kestle JR: The natural history of and tumors. Epilepsia 44:1413–1419, 2003
cerebral cavernous malformations. J Neurosurg 83:820–824, 55.  Xia C, Zhang R, Mao Y, Zhou L: Pediatric cavernous mal-
1995 formation in the central nervous system: report of 66 cases.
35.  Kraemer DL, Griebel ML, Lee N, Friedman AH, Radtke RA: Pediatr Neurosurg 45:105–113, 2009
Surgical outcome in patients with epilepsy with occult vascu- 56.  Yeon JY, Kim JS, Choi SJ, Seo DW, Hong SB, Hong SC: Su-
lar malformations treated with lesionectomy. Epilepsia 39: pratentorial cavernous angiomas presenting with seizures:
600–607, 1998 surgical outcomes in 60 consecutive patients. Seizure 18:14–
36.  McIntosh AM, Kalnins RM, Mitchell LA, Fabinyi GC, Bri- 20, 2009
ellmann RS, Berkovic SF: Temporal lobectomy: long-term 57.  Zevgaridis D, van Velthoven V, Ebeling U, Reulen HJ: Seizure
seizure outcome, late recurrence and risks for seizure recur- control following surgery in supratentorial cavernous malfor-
rence. Brain 127:2018–2030, 2004 mations: a retrospective study in 77 patients. Acta Neurochir
37.  Moran NF, Fish DR, Kitchen N, Shorvon S, Kendall BE, Ste- (Wien) 138:672–677, 1996
vens JM: Supratentorial cavernous haemangiomas and epilep-
sy: a review of the literature and case series. J Neurol Neu­
rosurg Psychiatry 66:561–568, 1999 Manuscript submitted March 28, 2011.
38.  Noto S, Fujii M, Akimura T, Imoto H, Nomura S, Kajiwara Accepted July 12, 2011.
K, et al: Management of patients with cavernous angiomas Please include this information when citing this paper: published
presenting epileptic seizures. Surg Neurol 64:495–499, 2005 online August 5, 2011; DOI: 10.3171/2011.7.JNS11536.
39.  Ojemann RG, Ogilvy CS: Microsurgical treatment of supra- Address correspondence to: Edward F. Chang, M.D., Department
tentorial cavernous malformations. Neurosurg Clin N Am of Neurological Surgery, University of California, San Francisco,
10:433–440, 1999 505 Parnassus Avenue, Box 0112, San Francisco, California 94143-
40.  Pozzati E, Acciarri N, Tognetti F, Marliani F, Giangaspero F: 0112. email: [email protected].

1174 J Neurosurg / Volume 115 / December 2011