SciELO - Scientific Electronic Library Online

vol.71 issue4Frequency of non-motor symptoms in Peruvian patients with Parkinson's diseaseProgrammable valve represents an efficient and safe tool in the treatment of idiopathic normal-pressure hydrocephalus patients author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Arquivos de Neuro-Psiquiatria

Print version ISSN 0004-282X

Arq. Neuro-Psiquiatr. vol.71 no.4 São Paulo Apr. 2013 


Intracranial cavernous malformation in children: a single-centered experience with 30 consecutive cases

Angiomas cavernosos intracranianos em crianças: experiência de um único centro em 30 casos consecutivos

Marcelo Campos Moraes Amato

João Flávio Gurjão Madureira

Ricardo Santos de Oliveira

1Division of Pediatric Neurosurgery, Department of Surgery and Anatomy, University Hospital of Ribeirão Preto Medical School, University of São Paulo, Ribeirão Preto, SP, Brazil



To determine the clinical presentation and treatment outcome of pediatric intracranial cavernous malformation (CM) in a single-centered institution.


Clinical data review of 30 patients under 18 years-old who had undergone surgery for cavernous malformation from January 1993 to December 2011.


The Study Group included 18 males and 12 females (mean age: 8.7 years-old). Symptoms at presentation were seizures (16/30, 53.3%), headache (15/30, 50.0%), and focal neurological deficits (11/30, 36.6%). Multiple cavernous malformations were found in 5/30 (16.6%). According to location, patients were classified in groups: (G1) brain-steam in 5/30 (16.6%), (G2) cerebellum in 2/30 (6.6%), (G3) supratentorial associated with seizures in 16/30 (53.3%), and (G4) supratentorial without seizures in 7/30 (23.3%). Surgical resection was performed in 26 out of 30 (86.6%) patients. The mean follow-up period was 4.1 years. Of 15 children followed-up with preoperative seizures, all were rendered seizure-free after surgery.


For symptomatic solitary cavernous malformation, the treatment of choice is complete microsurgical excision preceded by careful anatomical and functional evaluation. For multiple cavernous malformation or asymptomatic patients, the treatment modalities must be cautiously considered.

Key words: hemangioma; cavernous; seizure; brain tumor



Determinar a apresentação clínica e o acompanhamento do tratamento em crianças com angioma cavernoso intracraniano numa única instituição.


Revisão de dados clínicos de 30 pacientes menores de 18 anos com que passaram por uma cirurgia de angioma cavernoso intracraniano entre janeiro de 1993 a dezembro de 2011.


O grupo de estudo incluiu 18 sujeitos masculinos e 12 femininos (idade média: 8,7 anos). Os sintomas iniciais eram convulsões (16/30, 53,3%), cefaleia (15/30, 50,0%) e déficits neurológicos focais (11/30, 36,6%). Havia angiomas cavernosos intracranianos múltiplos em 5 de 30 (16.6%). A classificação foi feita em grupos de acordo com a localização: (G1) tronco cerebral em 5/30 (16,6%); (G2) cerebelo em 2/30 (6,6%); (G3) supratentoriais associados a convulsões em 16/30 (53,3%) e (G4) supratentoriais sem convulsões em 7/30 (23,3%). Ressecção cirúrgica foi realizada em 26 de 30 (86,6%) pacientes, com seguimento médio de 4,1 anos. De 15 crianças com convulsões pré-operatórias, todas ficaram livres das crises após a cirurgia.


Para angioma cavernoso intracraniano solitário e sintomático, o tratamento de escolha é excisão microcirúrgica total precedida de avaliação funcional e anatômica meticulosa. Para angiomas cavernosos intracranianos múltiplos ou pacientes assintomáticos, as modalidades terapêuticas devem ser consideradas cautelosamente.

Palavras-Chave: angioma cavernoso; convulsão; tumor cerebral

Intracranial cavernous malformations (CMs) are common vascular anomalies in the brain and have an incidence of 0.1 to 0.5% in the general population1.

These malformations represent 10 to 20% of all vascular lesions in the brain and can cause symptoms such as seizures, hemorrhages, headache, and focal neurological deficits2,3. The lesions are characterized by a “blackberry-like” aggregation of grossly enlarged capillary cavities consisting of a single layer of endothelium, without intervening neuronal tissue4. Since the introduction of sensitive diagnostic tools like the magnetic resonance imaging (MRI), CMs have become increasingly reported in the pertinent literature because asymptomatic lesions have also become detectable2,5.

Pediatric CMs may have characteristics that differ from those of adults in clinical presentation and treatment, such as higher rates of hemorrhage610 and larger dimensions9. Even though there are not many reports about exclusive pediatric CMs, only a few large series have yet been reported, and ideal group classification has not yet been defined.

This report is an attempt to evaluate clinical and surgical data in a single-centered pediatric series of CMs in the context of published literature. Special emphasis was placed on management of different groups and on outcome of patients suffering from seizure prior to surgery.


Patient population

We reviewed files of patients younger than 18 years harboring a CM, over an 18-year period at the Division of Pediatric Neurosurgery, University Hospital of Ribeirão Preto Medical School, from January 1993 to December 2011. Thirty CM cases were identified and enrolled in this study.

The variables analyzed included age, sex, clinical presentation, radiological features, extent of resection, and histopathology.

All patients had all been preoperatively evaluated by structural brain imaging with computed tomography (CT) and/or MRI. Shortly after surgical excision, all patients were put on regular follow-up.

The subjects were classified according to the main lesion location: brain-steam (G1), cerebellar (G2), supratentorial associated with seizures (G3), and supratentorial not associated with seizures (G4).

Postoperative seizure outcome was classified according to Engel's classification11: class I, free of disabling seizures; II, rare disabling seizures; III, worthwhile improvement; and IV, no worthwhile improvement.

A PubMed database search was also performed in order to retrieve articles published in the last 20 years, which combined the subject headings: cavernoma; cavernous angioma; pediatric; intracranial; and children. Papers with less than ten patients were excluded as well as series with patients above 20 years-old. Only articles in the English language containing all types of clinical manifestation of the CMs were considered, those concerning only seizures, brainstem, radiation-induced cavernomas or restricted types of cavernomas' classification were excluded from this review.

This study was approved by the Research Ethics Committee of the University Hospital of Ribeirão Preto Medical School, at University of São Paulo (Protocol 6591/2007).


Clinical and surgical data

The Study Group included 18 males and 12 females with a male-female ratio of 1.5:1. The mean age at onset ranged from 6 months to 17 years-old (mean 8.7 years-old). Patient's age and gender, CM location, clinical aspects, and outcome are summarized in Table 1.

Table 1. Summary of clinical-radiological characteristics and outcome of patients. 

Pt Age (years) Gender FH ML Localization of main lesion Main Symptoms Clinical Hemorrhage Group Surgery Technique Follow-up Outcome
1 6 M No Yes Pons Coma Yes G1 No 2 No deficit
2 14 M No No L cerebellar IH, ataxia Yes G2 Yes Lesionectomy 6 No deficit
3 0.6 F No No L thalamus Seizures, R hemiparesis Yes G3 Yes Lesionectomy 3 Seizure-free (Engel1a)
4 6 M No No R parietal Seizures No G3 Yes Lesionectomy 1 Seizure-free (Engel1a)
5 16 F No No Pons CN, hemiparesis Yes G1 Yes Lesionectomy 2 Transitory VI CN palsy
6 15 F No No L basal nuclei R hemiparesis Yes G4 No 6 No deficit
7 17 M Yes Yes L parietal Seizures No G3 Yes Lesionectomy + hemosiderin removal 13 Seizure-free (Engel1b)
8 5 M No No R frontal Seizures No G3 Yes Lesionectomy 5 Seizure-free (Engel1a)
9 4 M No No L frontal Seizures No G3 Yes Lesionectomy 6 Seizure-free (Engel1a)
10 12 F Yes Yes R Insula (multiple) Headache No G4 Yes Lesionectomy 2 Monoparesis
11 13 M No No R frontal Seizures No G3∗ Yes Lesionectomy + hemosiderin removal 1 Seizure-free (Engel1a)
12 1 M No Yes R occipital Seizures No G3 Yes Mesial occipital pole resection 2 Seizure-free (Engel1c)
13 1.25 M Yes No L temporal Seizures, hemiparesis Yes G3 Yes Left temporal pole resection 4 Seizure-free (Engel1b)
14 11 M No No L thalamus R hemiparesis Yes G4 No 2 No deficit
15 12 M No No L thalamus Headache Yes G4 Yes Lesionectomy 5 Epilepsy
16 14 F No No R frontal Seizures No G3 Yes Lesionectomy 2.33 Seizure-free (Engel1b)
17 16 M No No L frontal R hemiparesis Yes G4 Yes Lesionectomy 1 No deficit
18 12 M No No Medula oblongata Coma Yes G1 Yes Lesionectomy 5 No deficit
19 2 F No No Pineal Headache No G4 Yes Lesionectomy 12 No deficit
20 7 M No No Pons CN, hemiparesis Yes G1 Yes Lesionectomy 8 No deficit
21 14 F No No R frontal Seizures No G3∗ Yes Lesionectomy + hemosiderin removal 7 Seizure-free (Engel1a)
22 11 F No No Cerebellum Coma Yes G2 Yes Lesionectomy 9 No deficit
23 4 M Yes No R frontal Seizures No G3 Yes Lesionectomy 1 Seizure-free (Engel1a)
24 6 F Yes Yes L temporal Seizures Yes G3∗ Yes Lesionectomy + hemosiderin removal 3 Seizure-free (Engel1a)
25 6 F No No R occipital Headache Yes G4 Yes Lesionectomy 0.5 No deficit
26 2 F No No Pons Headache Yes G1 No 1 No deficit
27 11 F No No L temporal Seizures No G3∗ Yes Lesionectomy + amigdalo-hippocampectomy 6.5 Seizure-free (Engel1a)
28 11 M No No R temporal Seizures No G3∗ Yes Electrocorticography + lesionectomy 4 Seizure-free (Engel1b)
29 5 M No No L occipital Seizures Yes G3 Yes Lesionectomy 2 No deficit
30 8 M No No L temporal Seizures No G3 Yes Lesionectomy 1 No deficit

Pt: patient; ML: multiple lesions; FH: family history; M: male; F: female; R: right; L: left; CN: cranial nerve deficit; IH: intracranial hypertension; G1: brain-steam; G2: cerebellum; G3: supratentorial associated; G4: supratentorial; G3∗ patients with epilepsy that required further neurophysiologic assessment.

The most common presenting symptoms were seizures in 16/30 (53.3%), headaches in 15/30 (50%), followed by focal neurological deficits in 11/30 (36.6%), and behavior disturbance in 1/30 (3.3%). Acute intracranial hemorrhage was presented in 16/30 (53.3%).

Single CMs were found in 25/30 (83%), while five patients (17%) had multiple ones. Regarding only symptomatic lesions, 23 (76.6%) supratentorial and 7 (23.3%) infratentorial, location was as follows: frontal lobe (7/30, 23.3%), temporal lobe (6/30, 20.0%), occipital lobe (3/30, 10.0%), parietal lobe (1/30, 3.3%), insula (1, 3.3%), thalamus and basal nuclei (4/30, 13.3%), brainstem (5/30, 16.6%), and cerebellar hemisphere (2/30, 6.6%).

A total of 26 CMs undergone surgery and four patients were managed conservatively. There were no postoperative deaths or significant complications in this series. No child was asymptomatic prior to the resection.

In the G1 (5/30), three patients were submitted to surgery (Fig 1). Two patients that showed spontaneous total improvement did not undergone surgery. In the G2 (2/30), microsurgical resection was performed after cerebellar hemorrhage in both patients (Fig 2).

Fig 1. Pre (A) and postoperative (B) images of a brainstem cavernous malformation. This patient presented with tetraparesis and 11 points at the Glasgow Coma Scale. Three years after surgery she was asymptomatic. (A): axial T1-weighted magnetic resonance imaging revealed a typical cavernous malformation, characterized by heterogenic mass and hypointense halo signal around it; (B): sagittal T1-weighted gadolinium-enhanced magnetic resonance imaging revealed total removal of the cavernous angioma and decompression of brainstem (Patient 5). 

Fig 2. (A) brain computed tomography showed a hemorrhagic lesion on the left cerebellum hemisphere and incipient dilatation of ventricular temporal horns. Axial T1 (B) and T2-weighted (C) magnetic resonance imaging revealed heterogenic lesion with mass effect and fourth ventricle compression. Histopathology confirmed the diagnosis of a cavernous angioma (Patient 2). 

In G3 (16/30), 11 patients underwent simple lesionectomy without further neurophisiological evaluation. Five patients had long-lasting symptoms (more than one year) or high frequency of seizures and were therefore deeply investigated and treated (Table 1). In cases of epilepsy, the surrounding hemosiderin-stained tissue had also been removed. Of 15 children followed-up with preoperative seizures, all of them were rendered seizure-free (Engel's class 1) after the CM removal.

Seven patients composed G4. Five cases among them were submitted to surgery (Fig 3), and two were managed conservatively (thalamic and basal ganglia CM).

Fig 3. Axial T2-weighted magnetic resonance imaging (A) revealed a right occipital lesion with a hypointense halo around it, and other lesions can also be seen (arrows). The gradient-echo sequence (B) confirmed multiple cavernous malformations (Patient 12). 


The mean follow-up period was 4.1 years (range 6 months to 13 years). The overall post-treatment results were positive. Major morbidity and mortality from surgical procedures were absent. One patient (case 10) showed a permanent monoparesis after surgery.

Complete recovery of the neurological condition was achieved in 15 out of 17 (88.2%) children with neurological impairment prior to surgery (headache, focal deficits, and behavior disturbance).

Among the 15 patients with seizures, eight are now seizure — and drug-free; in seven patients, however, antiepileptic medications were not discontinued, although these children also scored Engel's class 1.

Four patients that were not operated improved spontaneously and remained asymptomatic in the follow-up.

During the follow-up period, no new lesions or recurrence on MRI images were seen in all children except for a boy (case 15) with a CM in the thalamus that had to be re-operated. He presented seizures due to severe hyponatremy after surgery and remained with antiepileptic drugs afterwards. The remaining lesions were stable in children with multiple CMs.


CM is one of the congenital vascular malformations affecting the central nervous system (CNS) that may remain asymptomatic for a long time or may produce clinical manifestations12. The pathogenesis of CM remains unclear, and little is known regarding the molecular pathogenesis of the disease3.

There are two types of CMs: sporadic and familial. The familial type is characterized by a multiplicity of lesions1315 and the possibility of de novo CM formation16. In this series, five patients (16.6%) had multiple lesions, three of them had also positive familial history. One subject with familial history (total of 4–13.3%) did not have multiple lesions. Among the literature review, incidence of familial history and multiple lesions both varies from 0.0 to 26.3%2,3,610,1719.

Table 2 summarizes previous studies on pediatric CM. Note that only ten series were according to the selection criteria and therefore included in this review2,3,610,12,17,19.

Table 2. Summary of literature review. 

Referência n Age Male:Female FH ML Site Main symptoms at diagnosis Post-surgical results
Supra:infratentorial Thalamus and Basal Gnaglia Brainstem Cerebellum Spine Seizures Headache Focal neurological deficits none Hemorrhage Surgery Follow-up Improved Unchanged Worsened Mortality
6 17 18 m to 16 y 1.12:1 4/17 (23.5%) 2/17 (11.7%) 15:2 2 2/17 (11.7%) 0 0 4 (23.5%) NA NA 0 12 (70.5%) 15 (88.2%) NA 11/15 (73.3%) 1/15 (6.6%) 3/15 (20%) 0
14 19 7 m to 17 y (mean=9,1) 1.11:1 5/19 (26.3%) 5/19 (26.3%) 12:6 2 4/19 (21%) 2/19 (10.5%) 1 5/18 (27.7%) 7/18 (38.8%) 13/18 (72.2%) 0 12/18 (66.6%) all 6 m to 9 y 17/19 (89.4%) 0 2/19 (10.5%) 0
18 18 10 m to 17 y 1:1.25 0 0 15:2 NA 0 2/18 (11.1%) 1 11/17 (64.7%) 1/17 (5.8%) 5/17 (29.4%) 0 NA all 1 to 16 y 14/17 (82.3%) 3/17 (17.6%) 0 0
7 24 6 m to 15 y (mean=8 y) NA 0 2/24 (8.3%) 20:4 NA NA NA 0 13 (54.1%) NA (9 IH) 10 (41.6%) 0 19 (79.1%) all 6 m to 14 y (mean=4 y) 23/24 (95.8%) 0 0 1/24 (4.1%)
8 36 9 m to 17 y (mean=8.6) 1.4:1 4/36 (11.1%) NA 23:12 6 7/36 (19.4%) 5/36 (13.8%) 1 16/35 (45.7%) 1/35 (2.8%) 10/35 (28.5%) 5/36 (13.8%) 19/36 (52.7%) all NA 30/35 (85.7%) 3/35 (8.5%) 2/35 (5.7%) 0
3 33 1 to 20 y (mean=11.6 y) 1.2:1 0 0 27:6 3 5/33 (15.1%) 1/33 (3%) 0 19 (57.6%) NA NA 0 25 (75.7%) 25 and RS on 8 2 to 17 y (mean=5.8 y) 29/33 (87.9%) 4/33 (12.1%) 0 1/33 (3%)
9 42 10 m to 17 y 1:1 1/42 (2.3%) 5/42 (11.9%) 35:5 0 2/42 (4.7%) 3/42 (7.1%) 2 28/40 (70%) 11/40 (27.5%) 16/40 (40%) 0 17/40 (42.5%) all 1 to 16 y 29/42 (69%) 10/42 (23.8%) 3/42 (7.14%) 0
19 66 15 m to 17.8 y (mean=11.6 y) 1.53:1 1/66 (1.5%) 7/66 (10.6%) 55:6 (both:4) 1 2/66 (3%) 4/66 (6%) 1 31/65 (47,7%) 30/65 (46,2%) 8/65 (12,3%) 2/65 (3,1%) 13/65 (20%) 62/66 (93.9%) 5 m to 9.3 y (mean=3.2 y) 43/46 (93.5%) 1/46 (2.2%) 2/46 (4.3%) 0
10 32 2 d to 17 y (mean=7.1) 1.13:1 3/32 (9.3%) 8/32 (25%) 24:08 1 6/32 (18.7%) 2/32 (6.2%) 0 13 (40.6%) 6 (18.8%) 7 (21.8%) 0 21 (65.6%) 28 (87.5%) mean=4.43 y 27/28 (96.4%) 1/28 (3.5%) 0 0
2 79 4 m to 17 y (mean=9.7 y) 1.07:1 1/79 (1.2%) 3/79 (3.8%) >74:NA NA NA NA 0 41 (51%) 8 (10%) 14 (18%) 0 18 (23%) all 1 m to 16 y (mean=3 y) 39/54 (72%) 12/54 (22.2%) 1/54 (1.8%) 0

n: number of patients; ML: multiple lesions; FH: family history; NA: not available; IH: intracranial hypertension; RS: radiosurgery; >: at least. Data were taken from the text and tables presented in the articles. Symptoms and post-surgical results excluded sometimes the spinal cord cavernomas and/or the patients without enough follow-up, correct amounts of patients are shown.

In our series, the mean age at clinical manifestation was 8.7 years-old, which is similar to the literature. The ratio of male to female was 1.5:1. In addition to the present study, there seems to be a slightly predominance of boys in most analyzed studies (Table 2). Since these are patient-based studies, the definitive predominance is hard to define.

The majority of CMs come to medical attention after children experience any number of clinical symptoms, and up to 20% are discovered incidentally1921. In pediatric patients, CM symptoms in the CNS are mainly present as hemorrhage, seizure, or focal neurological deficits3. Xia et al. also observed acute intracranial hemorrhage in 20% of children with CMs, according to CT and MRI studies. However, a higher proportion of hemorrhage was confirmed histopathologically in 52.5% (32/61) of operated children with CMs in their series19. Acute intracranial hemorrhage was observed in 53.3% (16/30) in our series.

In the current study, 56.6% of patients suffered from seizures, which is significantly higher than in earlier studies2,68,10,17,19, although Acciarri et al.9 reported 70% of seizures in a series of 42 children with CMs.

Presentation of symptoms has some correlation with lesion locations19. Patients with lesions located superficially in the cerebral hemispheres more frequently are present with seizures. On the other hand, patients with lesions located in deep structures such as the brainstem generally present with focal neurological deficits. In this series, focal neurological deficits (36.6%, 11/30) happened mostly in deep-seated and infratentorial lesions. Only two cases with focal neurological deficits did not result from deep-seated lesions, an acute hemorrhage of a temporal (case 13) and a frontal CM (case 17).

CMs have been observed throughout the CNS, and their locations in children are comparable to those in adults8,19,22. Supratentorial location accounts for about 80.0%, while the other 20.0% are located in the posterior cerebral fossa3,8,19,22. The proportion of supratentorial CMs in the descending order is frontal, temporal, occipital, and parietal lobe. A deep location in the basal ganglia, hypothalamus, or ventricular system is rare8,19,22. In this series, regarding only the symptomatic lesions, 23 patients (76.6%) presented supratentorial CM and seven (23.3%) infratentorial.


Presence or absence of hemorrhage is important to decide how to manage the patients. The risk of bleeding is higher after the first event, 22.9% against 0.25–3.8% without prior hemorrhage4. However, without a documented episode of hemorrhage, treatment should be based on the precise location of the cavernoma, patient's age, symptoms, and outcome expectation23,24.

Additional data of natural history also favor surgical treatment. Seizures seem to happen precociously when CMs are located in the frontal or temporal lobes25, and apparently their risk of development varies from 1.5 to 4.8% per year4,21. Focal deficits and intracranial hypertension may be caused not only by acute macro-hemorrhage, but also due to progressive growing secondary to recurrent micro-hemorrhage24.

Figs 4 and 5 summarize management strategies for each following group.

Fig 4. Therapeutic guideline for G1 and G2 (modified from Samii et al.23). 

Fig 5. Therapeutic guideline for G3 and G4. 

Brain steam cavernomas (G1)

Brainstem CMs constitute 0 to 21% of all cavernomas in different pediatric series3,6,810,1719. Clinical presentation is generally acute and severe due to peculiar localization, despite frequent small amount of hemorrhage. Neurological improvement often happens a few days later. This apparently benign course is due to slow flow and pressure inside the CMs, consequently hemorrhage usually displaces the neural tissue instead of its destruction7. Two of our patients were not submitted to surgery and improved totally spontaneously.

Surgical resection is indicated for those brainstem lesions that appear in the pial surface23. Asymptomatic patients or with good neurological recover should be observed, mainly if the CMs are located deeply in the brain steam4,17. Some authors believe that the risk of not operating on a patient that had already suffered hemorrhage is very high, especially if he/she is young with high life expectancy, independent of the lesions depth, because of the possible catastrophic evaluation if it rebleeds or grows3,26. Neurosurgical team experience and skills, as well as neurophysiologic monitoring, should be considered to treat such patients19,23.

Cerebellar cavernomas (G2)

Cerebellar location is less frequent, and clinical presentation is generally due to great hemorrhage inside the cerebellar parenchyma that may cause mass effect and occlude the fourth ventricle resulting in cerebellar dysfunction and hydrocephalus27.

Patients with cerebellar cavernoma (G2) should be treated with surgery even in the absence of hemorrhage. Surgery is usually safe and successful while a first or recurrent hemorrhage in the posterior fossa represents great threat. Exception is made for multiple asymptomatic cerebellar CMs4. Cerebellar cavernomas represent 0.0 to 13.8% in earlier studies3,6,810,1719, in our series there were two patients in this group (6.6%).

Supratentorial cavernomas associated with seizure (G3)

Surgical treatment should be indicated to the majority of patients that suffered hemorrhage and to those which surgical risk was considered low4,9.

Lesionectomy alone showed favorable results28. In a series of patients submitted to lesionectomy of supratentorial CMs, a strict relationship was found between the duration of seizure history before the surgical treatment and the cure of epilepsy with the surgery. Cohen et al.28 showed that all patients that presented one preoperative seizure or less than two months of history did not have any seizures after surgery; 75.0 to 80.0% of patients with two to five preoperative seizures or 2 to 12 months of history also had not any seizure after surgery. In addition, 50% of patients with more than five seizures or more than 12 months of history were postoperatively seizure free28. Although resection of the hemosiderin stained tissue that surrounds the lesion was not proved to be of better effectiveness4,810,29, it was adopted by several studies when CMs are placed in noncritical areas2,3,9, including the present one. Eleven patients of this study underwent simple lesionectomy.

Longer seizure history and increased number of preoperative seizures are consistent with more difficulty in treating epilepsy with simple lesionectomy surgery28. We believe these patients should have a complete epilepsy evaluation to a possible extended surgery guided by electrocorticography. In fact, five patients with refractory seizures were further evaluated, and two of them needed an extended resection.

This group represented 53.3% of patients (16/30) and surgical techniques chosen for each case are showed in Table 1.

Supratentorial cavernomas not associated with seizure (G4)

This group represented 23.3% of patients (7/30). The management of thalamic and basal ganglia CMs is controversial4. In selected cases, surgery is considered when the CM is next to the ventricle surface or in the insular lobe through to the Sylvian fissure approach10. Indeed, among three patients, two were conservatively treated with a remarkable neurological outcome.

Recurrent headache seems to be attributed to the CM. Surgery is indicated to prevent future neurologic deficit, except for patients with unacceptable risk due to critical lesion location4,9. In this series, one patient (case 19) with a pineal lesion had insidious onset of headache and somnolence. Another patient (case 10) with multiple lesions had history of recurrent headache and behavior abnormality, the insular lesion was operated through a pterional approach and a monoparesis persisted in the follow-up.

If a focal neurologic deficit is present, there is usually an associated hematoma, and the surgical removal should be performed in order to avoid further deficit and to help clinical recovery. The management of CM in eloquent brain areas (motor, sensitive, speech, or visual) should be tailored for each case4.

In conclusion, we found a favorable outcome for surgically treated children with symptomatic CMs, and we suggest that resection should be the golden-standard therapy for patients with lesions that do not cause an excessive surgical risk, such as localization in a highly eloquent area.

Natural history is particularly important in children and should be individualized, considering besides bleeding also CM location, symptoms, surgical treatability, and life expectancy. Surgical removal of CMs in eloquent areas, such as brainstem and thalamus/basal nuclei, can prevent neurological deterioration secondary to re-bleeding or more seldom lesion growth, but these are still challenging and conservative treatment is still an option, particularly in those cases with considerable clinical improvement. Cerebellar cavernoma should be generally treated with surgery. Recent record of seizures is associated with high rate of success after lesionectomy. Extended surgery associated with CM lesionectomy for long-term epilepsy or high frequency of seizures should be considered after proper investigation and failure of drug therapy.

Conservative treatment may be considered for asymptomatic patients, multiple CMs, critical lesion location (brainstem, thalamus, and basal nuclei) and in cases of small lesions that are not associated with hemorrhage.


1. Del Curling O, Kelly DL, Elster AD, Craven TE. An analysis of the natural history of cavernous angiomas. J Neurosurg 1991;75:702-708. [ Links ]

2. Hugelshofer M, Acciarri N, Sure U, et al. Effective surgical treatment of cerebral cavernous malformations: a multicenter study of 79 pediatric patients. J Neurosurg Pediatr 2011;8:522-525. [ Links ]

3. Lee JW, Kim DS, Shim KW, et al. Management of intracranial cavernous malformation in pediatric patients. Childs Nerv Syst 2008;24:321-327. [ Links ]

4. Amin-Hanjani S, Ojemann RG, Ogilvy CS. Surgical management of cavernous malformations of the nervous system. In: Schmidek HH (Ed). Operative neurosurgical techniques: indications, methods and results. Philadelphia: Elsevier, Inc; 2006. p. 1307-1324. [ Links ]

5. Faria MP, Fagundes-Pereyra WJ, Carvalho GT, Sousa AA. Cavernous malformations: surgical management in Belo Horizonte Santa Casa Hospital. Arq Neuropsiquiatr 2004;62:1079-1084. [ Links ]

6. Mazza C, Scienza R, Beltramello A, Da Pian R. Cerebral cavernous malformations (cavernomas) in the pediatric age-group. Childs Nerv Syst 1991;7:139-146. [ Links ]

7. Di Rocco C, Iannelli A, Tamburrini G. Surgical management of paediatric cerebral cavernomas. J Neurosurg Sci 1997;41:343-347. [ Links ]

8. Mottolese C, Hermier M, Stan H, et al. Central nervous system cavernomas in the pediatric age group. Neurosurg Rev 2001;24:55-71. [ Links ]

9. Acciarri N, Galassi E, Giulioni M, et al. Cavernous malformations of the central nervous system in the pediatric age group. Pediatr Neurosurg 2009;45:81-104. [ Links ]

10. Consales A, Piatelli G, Ravegnani M, et al. Treatment and outcome of children with cerebral cavernomas: a survey on 32 patients. Neurol Sci 2010;31:117-123. [ Links ]

11. Engel JJ, Van Ness PC, Rasmussen TB, Ojemann LM. Outcome with respect to epileptic seizures. In: Engel JJ (Ed). Surgical treatment of the epilepsies. New York: Raven Press; 1993. p. 609-621. [ Links ]

12. Giulioni M, Acciarri N, Padovani R, Galassi E. Results of surgery in children with cerebral cavernous angiomas causing epilepsy. Br J Neurosurg 1995;9:135-141. [ Links ]

13. Rigamonti D, Spetzler RF. The association of venous and cavernous malformations. Report of four cases and discussion of the pathophysiological, diagnostic, and therapeutic implications. Acta Neurochir (Wien) 1988;92:100-105. [ Links ]

14. Scott RM. Brain stem cavernous angiomas in children. Pediatr Neurosurg 1990;16:281-286. [ Links ]

15. Zabramski JM, Wascher TM, Spetzler RF, et al. The natural history of familial cavernous malformations: results of an ongoing study. J Neurosurg 1994;80:422-432. [ Links ]

16. Tekkök IH, Ventureyra EC. De novo familial cavernous malformation presenting with hemorrhage 12.5 years after the initial hemorrhagic Ictus: natural history of an infantile form. Pediatr Neurosurg 1996;25:151-155. [ Links ]

17. Scott RM, Barnes P, Kupsky W, Adelman LS. Cavernous angiomas of the central nervous system in children. J Neurosurg 1992;76:38-46. [ Links ]

18. Giulioni M, Acciarri N, Padovani R, Frank F, Galassi E, Gaist G. Surgical management of cavernous angiomas in children. Surg Neurol 1994;42:194-199. [ Links ]

19. Xia C, Zhang R, Mao Y, Zhou L. Pediatric cavernous malformation in the central nervous system: report of 66 cases. Pediatr Neurosurg 2009;45:105-113. [ Links ]

20. Robinson JR, Awad IA, Little JR. Natural history of the cavernous angioma. J Neurosurg 1991;75:709-714. [ Links ]

21. Pozzati E, Padovani R, Morrone B, Finizio F, Gaist G. Cerebral cavernous angiomas in children. J Neurosurg 1980;53:826-832. [ Links ]

22. Lena G, Ternier J, Paz-Paredes A, Scavarda D. Central nervous system cavernomas in children. Neurochirurgie 2007;53:223-237. [ Links ]

23. Samii M, Eghbal R, Carvalho GA, Matthies C. Surgical management of brainstem cavernomas. J Neurosurg 2001;95:825-832. [ Links ]

24. Detwiler PW, Porter RW, Zabramski JM, Spetzler RF. De novo formation of a central nervous system cavernous malformation: implications for predicting risk of hemorrhage. Case report and review of the literature. J Neurosurg 1997;87:629-632. [ Links ]

25. Requena I, Arias M, López-Ibor L, et al. Cavernomas of the central nervous system: clinical and neuroimaging manifestations in 47 patients. J Neurol Neurosurg Psychiatry 1991;54:590-594. [ Links ]

26. Di Rocco C, Iannelli A, Tamburrini G. Cavernous angiomas of the brain stem in children. Pediatr Neurosurg 1997;27:92-99. [ Links ]

27. Cavalheiro S, Braga FM. Cavernous Hemangioma. In: Choux M, Di Rocco C, Hockley A, Walker M (Eds). Pediatric neurosurgery. London: Churchill Livingstone; 1999. p. 691-701. [ Links ]

28. Cohen DS, Zubay GP, Goodman RR. Seizure outcome after lesionectomy for cavernous malformations. J Neurosurg 1995;83:237-242. [ Links ]

29. Di Rocco C, Iannelli A, Tamburrini G. Cavernomas of the central nervous system in children. A report of 22 cases. Acta Neurochir (Wien) 1996;138:1267-1274. [ Links ]

Received: May 8, 2012; Received: October 19, 2012; Accepted: October 26, 2012

Correspondence: Marcelo Campos Moraes Amato; Avenida Juriti 144; 04520-000 São Paulo SP - Brasil; E-mail:

Conflict of interest: There is no conflict of interest to declare.

Creative Commons License This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.