|Year : 2020 | Volume
| Issue : 1 | Page : 32-39
Management strategies for cerebral arteriovenous malformation - An institutional experience
Kasinathan Sudhakar1, Abderrahmane Cheikh2, Yashiro Yamada3, Takao Teranishi2, Tsukasa Kawase2, Yoko Kato2
1 Institute of Neurosurgery, Madras Medical College and Government Rajiv Gandhi General Hospital, Chennai, Tamil Nadu, India
2 Department of Neurosurgery, Ali Ait Idir Hospital Algiers, Algeria
3 Department of Neurosurgery, Fujita Health University, Banbuntane Hotokukai Hospital, Nagoya, Japan
|Date of Submission||06-Sep-2020|
|Date of Acceptance||01-Sep-2020|
|Date of Web Publication||1-Oct-2020|
Dr. Kasinathan Sudhakar
Institute of Neurosurgery, Madras Medical College and Government Rajiv Gandhi General Hospital, Chennai, Tamil Nadu
Source of Support: None, Conflict of Interest: None
Background: Cerebral Arteriovenous Malformation represent a particularly challenging subset of lesions, given their overall low incidence and the high frequency with which some form of multimodality monitoring may be required for complete obliteration.
Materials and Methods: We report retrospective analysis of 9 cases with unruptured cerebral arteriovenous malformation who got admitted in the Banbuntane Hotokokai Hospital, Fujita Health University, and Nagoya, Japan .from the year 2014 to 2017.surgical treatment was appropriate craniotomy and resection based on the location of the arteriovenous malformation with multimodality monitoring with Dual Image Video Angiography, Motor Evoked Potential, and FLOW 800. We studied the following factors: Sex, Age, Arteriovenous Malformation size, Location, occurrence of Intracranial Hemorrhage, Seizure type, Duration of Seizure history, Treatment modality, and Arteriovenous Malformation obliteration. We tested for statistical associations between these factors and seizure presentation and outcomes with Clinical follow-up. Outcome was compared with modified Rankin Scale
Results: Out of 9 cases operated for intracranial arteriovenous malformation 5 patients were female and 4 were male. Age [ Mean 39.4 + 16.7] of the patients ranged from 20 to 66 , 5 patients were 25 - 40 and 2 were > 65 and < 25 each.6 patients were Spetzler - Martin Grade 2 ,1 patient was Spetzler - Martin Grade 4,Grade 3 and Grade 5. 3 patients were draining into superficial venous system and 2 patient were asymptomatic and incidentally diagnosed during screening procedure. There is no Mortality in this study.1 patient underwent Gamma Knife stereotactic radiosurgery.
Conclusion: The increasing use of advance imaging techniques will increase the incidence of asymptomatic arteriovenous malformations. Non-ruptured arteriovenous malformation Spetzler Martin Grade 1 or Spetzler Martin Grade 2 have a good outcome for microsurgery [modified Rankin Scale and complete obliteration and there is better outcome with microsurgery.
Keywords: Cerebral Arterio venous malformation, A Randomized Trial of Unruptured Brain Arteriovenous Malformations, Spetzler Martin Grade, unruptured arterio venous malformation, Microsurgery, modified Rankin Scale
|How to cite this article:|
Sudhakar K, Cheikh A, Yamada Y, Teranishi T, Kawase T, Kato Y. Management strategies for cerebral arteriovenous malformation - An institutional experience. J Cerebrovasc Sci 2020;8:32-9
|How to cite this URL:|
Sudhakar K, Cheikh A, Yamada Y, Teranishi T, Kawase T, Kato Y. Management strategies for cerebral arteriovenous malformation - An institutional experience. J Cerebrovasc Sci [serial online] 2020 [cited 2020 Nov 28];8:32-9. Available from: http://www.jcvs.com/text.asp?2020/8/1/32/296937
| Introduction|| |
Brain arteriovenous malformations (AVMs) are abnormal connections between arteries and veins, leading to arteriovenous shunting with an intervening network of vessels – the so-called nidus. They are incidental findings in 0.05% with overall detection rate being 1 per 100,000 adults per year, and they have accepted bleed risk of 2%–4% per year with an associated neurological morbidity of 20%–30% and mortality of 10%–30% with each bleed. Roughly half of patients with brain AVMs present with intra-cranial haemorrhage. This article gains importance from the controversy regarding the management of unruptured AVMs, particularly aftermath of A Randomised Trial of Unruptured Brain Arteriovenous Malformations (ARUBA) study.
| Subjects and Methods|| |
We performed a retrospective analysis of 10 medical records of patients at Bantane Hospital, Nagoya, Japan, who were diagnosed to have brain AVM and subsequently managed from February 2015 to November 2017. The pre- and post-treatment radiographic studies and clinical follow-up evaluations were reviewed retrospectively. Since patient case record was only used and the identity was masked, consent for the study was not required.
Spetzler Martin grade
Spetzler Martin grade (SMG) scale, which is a composite score of: nidus size (<3 cm, 3–6 cm and > 6 cm: 1–3 points); eloquence of adjacent brain (1 point if located in brainstem, thalamus, hypothalamus, cerebellar peduncles or sensorimotor, language or primary visual cortex); presence of deep venous drainage (1 point if any or all drainage is through deep veins, such as internal cerebral veins, basal veins or pre central cerebellar veins).
Supplementary Spetzler Martin grade
Lawton et al. introduced a supplementary grading system including additional variables such as patient age, haemorrhagic presentation, nidal diffuseness and deep perforating arterial supply. The supplementary grading system is a better predictor of neurologic outcomes after AVM surgery. Supplementary SMG of 6 is a cut-off or boundary for the decision of microsurgery.
A Randomised Trial of Unruptured Brain Arteriovenous Malformations
A Randomised Trial of Unruptured Brain Arteriovenous Malformations (ARUBA) study compared medical management alone with medical management plus interventional therapy. Majority of the patients were identified with SMG I, II or III with a few with grade IV (10%), and no patients with grade V were enrolled. ARUBA test is the beginning of the optimal treatment for the unruptured AVMs rather than the definitive word. Other similar experiments, such as the Scottish Intracranial Vascular Malformations Study (SIVMS), reached conclusions similar to those of the ARUBA trial, and the present scenario in our study is as follows:
- Non-rupture does not entail a lack of symptoms. For patients with epilepsy or focal neurological deficit, active treatment is recommended on the condition that the treatment risk is under control
- It is not advisable to simply conclude from the results of the ARUBA and SIVMS tests that no treatments are needed for non-ruptured AVMs, but instead that the natural risk of the disease and the results after treatment should be balanced. No active treatment is recommended because no choices are optimal. Hence, treatment plans were optimised for the presentation of unruptured AVMs based on the SMG and supplementary SMG.
Endovascular embolisation was done as either pre- or intra-operative embolisation with Onyx, which attempts to occlude the nidus and feeding vessels as a curative procedure, palliative flow reduction, targeting of specific angio-architecture features, preoperative flow reduction or pre-radiosurgical volume reduction.
Microsurgery for resecting the AVM itself, and/or in combination with endovascular embolisation or coiling by appropriate craniotomy based on the location, and AVM resection alone or with aneurysm clipping if related to AVM. Emergency surgery should be considered for a patient when a rapidly declining neurological status is attributed to a ruptured AVM.
Microsurgical adjuncts in high-end operation theatre
ICG is a rapidly cleared, fluorescent, hydrophilic tri-carbo-cyanine molecule that is ideal for angiography. With an intact blood–brain barrier, the dye remains within the cerebral vasculature. It is evident in the arterial phase within 3–12 s of a venous injection. Its absorption and emission peaks are 805 and 835 nm, respectively. The real benefit of ICG angiography is that it provides rapidly acquired information that is immediately integrated in the surgical view. The technique is ideal for the early identification of AVM arteries and veins, helping a surgeon to formulate and modify an operative strategy for attacking these formidable lesions.
Dual-image video angiography
Dual image video angiography, a high resolution, intra operative imaging system, simultaneously visualises both visible light images and near infrared fluorescence images from ICG. Dual image video angiography is used in identifying the feeder and drainer and passing through artery and for confirmation of residual AVM and intra operative gradual nidus occlusion.
FLOW 800 software allows objective time dependent assessment of intra operative blood flow during the removal of AVMs, and it mixes ICG VA data with a colour coded map that helps determine flow direction, and flow speed, and visually separate feeding arteries from arterialised veins in AVM surgery.,
Gamma knife radiosurgery involves the targeting of the AVM nidus and adjacent vessels intended to induce a reduction in AVM size and possible obliteration of the AVM. Single-session SRS is not effective for high-grade AVM. Hence, volume-staged SRS is a newer strategy which divides a large AVM into 2–3 smaller portions, with a higher radiation dose which increases obliteration rate. Post-radiosurgical AVMs are more favourable for microsurgery as they are amenable to resection.
The primary outcome measure was functional outcome at last, follow up based on the modified Rankin scale (mRS) score dichotomised to ”good” (mRS 0–1) or ”poor” (mRS 2–6). AVM obliteration was documented by catheter angiography.
Statistical analysis was performed with Python language tool, and analyses of continuous variables were presented as means ± standard deviation and categorical variables as frequencies (percentages). Demography including sex, age, clinical presentation and morphology characteristics such as location, eloquence, venous drainage, size, SM score and SM-supplemented score were compared with complications and outcomes with two-tailored test (Levene's test[mean]) for variance, and a P < 0.05 was considered statistically significant.
| Results|| |
Demography and arteriovenous malformation characteristics
Of 10 patients whose medical records were analysed and follow-up was done, five patients were aged 25–45 years, 1 patient aged 40–65 years and 2 patients were aged <25 and >65 years each.
There were 5 (50%) male and 5 (50%) female patients [Table 1].
SMG and supplementary SMG were used to characterise the brain AVMs. The most common size was 3–6 cm and found in 5 (55.5%) patients. 3 (33.3%) patients had AVM size <3 cm while 1 (1.1%) patient had size >6 cm. 5 (55.5%) patients had deep venous drainage system while 2 (2.2%) patients were in eloquent region, 2 (2.2%) patients were diffuse in architecture while 1 (1.1%) patient had deep perforator supply and 6 (60%) patients presented with ruptured AVM to the hospital. 1 (10%) patient had intra-nidal aneurysm associated with AVM and one patient detail was not traceable.
Cerebral AVMs were mainly located in the lobar region in which frontal, temporal, and occipital being the most common in 2 (2.2%) patients each. 1 (1.1%) patient each had AVM in the thalamic and cerebellar region. 1 (1.1%) patient had large-sized AVM extending from the frontal to occipital region [Chart 1].
Incidental ARUBA eligible presentation is most common in 4 (40%) patients. While 2 (20%) patients presented with multiple episodes of intra-cerebral haemorrhagic manifestations, 1 (10%) patient each presented with epilepsy, loss of consciousness and haemorrhage in the subdural, subarachnoid and intra-ventricular region. 4 (40%) patients had pre-operative mRS as 0, while 3 (30%) patients with mRS as 3, 2 (20%) patients with mRS as 5 and 1 (10%) had mRS as 1 [Chart 2].
Spetzler Martin grade
Grade I and II (66.6%) are the most common presentations with SMG S1 V0 E1 and S1 V0 E0 in 1 patient each; S1 V1 E0 and S2 V0 E0 in 2 (22.22%) patients each; and 2 (22.22%) patients with grade IV in S3 V0 E1 and S2 V1 E1 combination. 1 patient (9%) presented with grade III pattern as S2 V1 E0 [Chart 3].
Supplementary Spetzler Martin grade
Four (44.44%) patients presented with grade IV while 4 (44.44%) patients presented with grade III and 1 (9%) patient presented with grade II [Chart 4].
Out of 10 patients, 1 (10%) patient underwent emergency surgery after rupture of the AVM and the post-operative details were not traceable. 2 (20%) patients did not undergo pre-operative embolisation while 2 (20%) patients underwent one-time pre-operative embolisation, 3 (30%) patients underwent two-time pre-operative embolisation and 2 (20%) patients underwent four-time pre-operative embolisation. 6 (60%) patients underwent only microsurgical resection of the AVM. While 3 (30%) patients underwent microsurgery after coiling of the AVM as it was partially obliterated, 1 (10%) patient underwent gamma knife radiosurgery following coiling. All the patients (100%) who underwent microsurgery had complete obliteration of the AVM which was confirmed by post-operative digital subtraction angiography [Chart 5] and [Flowchart 1].
The overall complication rate was 30% in microsurgery in which two patients developed intra-cranial haemorrhagic manifestation after post-operative embolisation for which one patient was operated. One patient developed post-operative intra-cranial hypertension, for which decompressive craniectomy was done and 1 (10%) patient was partially obliterated after gamma knife radiosurgery, for which gamma knife revision was done.
Outcome postoperative modified Rankin Scale
Out of the nine patients for which post-operative data are available, 4 (45%) patients improved while 3 (33%) patients' post-operative mRS was unchanged when compared with the pre-operative mRS and 2 (22%) patients worsened. There was no mortality in our study.
According to analysis [Graph 1] of available data with two tailored (Levene's test[mean]) test, we get the outcome as rejecting the null hypothesis that endovascular management and microsurgical management gives the similar outcome; we obtained a P < 0.0001 which is statistically significant [Chart 6].
As the computed P value is lower than the significance level alpha = 0.05, one should reject the null hypothesis H0 and accept the alternative hypothesis Ha.
Test interpretation is given below:
- H0: The variances are identical
- Ha: At least one of the variances is different from another.
| Discussion|| |
Baseline clinical presentation, AVM characteristics and demography are comparable to available literature. In Lawton et al.'s study, the average age of the AVM population from the study was 37 years, which is similar to ours (39.4 years). In Wong et al.'s study, there were 57% females and 43% males, which is similar to our study. Most common location of AVMs are frontal (29.36%), parietal (27.7%) and temporal (25.2%), which is similar to our study showing 20% in each region. Our study had 10 patients out of which 9 (90%) patients received microsurgery either alone (30%) or in combination with endovascular surgery (60%), and 1 (10%) patient underwent gamma knife radiosurgery following endovascular procedure. The mean duration of follow-up in our series is 33 months, which is the same as the mean follow-up in ARUBA trial. In our study, younger age was a good prognostic factor influencing the outcome.
For cerebral AVM, a considerable body of evidence exists to support treatment in many patients, even when asymptomatic. We had four patients with incidental presentation of which three were SMG II and one with SMG III in which one patient with SMG II had poor outcome (mRS > 2) and three patients had good outcome (mRS 0).
A Randomised Trial of Unruptured Brain Arteriovenous Malformations eligible
In our study, ARUBA eligible patients were 4 (40%) of which majority (75%) were of SMG II. ARUBA confirmed what we already knew – that is, leaving an AVM untreated brings a high lifetime stroke risk. According to the New York Islands AVM Study, there are three identified factors associated with an increased risk of rupture of an AVM (previous rupture, deep venous drainage and deep location), leading to a reported annual risk from 0.9%–32%. Our ARUBA eligible patients had excellent outcome (mRS 0) in 75% of the patients which is contrary to the available literature which can be explained by the SMG of the patient; most of them were grade I and II.
In Lawton et al.'s study, after their initial haemorrhage, the majority of patients (72%) had a mRS of >2 and 61% of patients had NIHSS scores of 0 or 1, which is comparable to our study in which all (100%) the patients with initial haemorrhage had poor outcome (mRS of >2). Lawton et al. state that haemorrhage was positively associated with SMG I and negatively with grade V which is very much similar to our study. AVMs with small nidus, single feeding artery and single draining vein are associated to haemorrhagic presentation, which endorses our study.
Seizure and arteriovenous malformation
AVMs are initially manifested by seizure in 15%–47% of the cases, which is similar to our study (10%). Turjman et al. identified six angio-architectural AVM factors associated to seizure: cortical location, feeding by the middle cerebral artery, feeding by cortical artery, an absence of aneurysms, the presence of varices and an absence of intra-nidal aneurysms were significantly associated with seizures. The relation between seizure and large nidus size was positive however negative for small nidus size. If the AVM was obliterated, there was no significant difference in outcome between patients who underwent surgical extirpation and those who underwent radiosurgical obliteration with or without adjunctive embolisation, which is similar to our study; the patient had good outcome; post operative mRS score was 0.
Focal neurological deficit
Progressive neurological dysfunction in AVM patients, including cognitive deficit, is usually related to hypo-perfusion and consequent brain ischemia. It can also be related, in some cases, to small recurrent haemorrhages, mass effect or hydrocephalus. In our study, focal neurological deficit in unruptured AVMs was 10%, similar to the literature. Risk factors associated are absence of aneurysm, presence of venous ectasia and association of ectatic vein in the absence of arterial aneurysm
Spetzler Martin Grade I And II Arteriovenous Malformations
Surgery has been claimed to be the preferred therapeutic option for all patients with grade I and II brain AVMs because of both the low morbidity, which is very much similar to our study. The nature of grade I and II AVMs is that if they are less than 3 cm in maximum diameter, they may be located in eloquent cortex or have deep venous drainage, but not both. In the absence of either deep venous drainage or a location in eloquent cortex, they must be less than 6 cm in maximum diameter. Therefore, the relatively uncomplicated nature of these AVMs accounts for the low morbidity associated with their surgical treatment. This has led to reported morbidity and mortality rates as low as 0% cure rate, which endorses our outcome of 0% mortality. However, the low management risk and the high cure rate in our microsurgical series reinforce the need for grade I and II AVMs to be considered for surgery first, and only those few cases who are considered high-risk grade I and II AVMs should be referred for focused irradiation. In this series, the greater than 95% chance of an excellent surgical outcome for patients with eloquent grade II AVMs and a greater than 99% chance of an excellent surgical outcome for patients with non-eloquent grade I or II AVMs continue to bolster the argument for surgical management of low-grade AVMs, which is similar to our findings of 100% cure in Grade II in eloquent AVMs combined with the opportunity to obtain an immediate angiographically demonstrated cure; this low risk still renders surgery preferable to radiosurgery for the treatment of these lesions. The findings in our study also corroborate that SMG I and II have better outcome in microsurgical management.
Spetzler Martin grade III arteriovenous malformations
Lawton alone and with colleagues reported that age, AVM size, unruptured presentation and diffuse nidus were significant predictors of outcome (measured by worsening of mRS scores), whereas eloquence and deep venous drainage were borderline predictors and deep perforating arterial supply did not predict outcome. The size of the AVM was critically predictive of outcome. Hence, we believe that grade III AVMs should be predominantly sub-classified into grade III (small) and grade III (large). Lawton described an obliteration rate of 97.4% with surgery in grade III AVMs and a procedure–ailure rate of 5.3%. Grade III AVMs (S2V1E0) require carefully individualised treatment recommendations and planning. However, our case was S2 V1 E0 which was managed by preoperative embolisation, coiling and microsurgery, and the patient had good outcome (mRS 0) [Figure 1] and [Figure 2].
|Figure 1: Illustrative case 1 Spetzler Martin grade III right temporal arteriovenous malformation. (a) Pre-operative three-dimensional CTA, (b) digital subtraction angiography, (c) intra-operative picture, (d) ICG-VA|
Click here to view
|Figure 2: Illustrative case 2 Spetzler Martin grade III left frontal arteriovenous malformation. Microsurgery done after endovascular coiling. (a) Three-dimensional CTA of arteriovenous malformation, (b) preoperative digital subtraction angiography, (c) intra-operative picture, (d) ICG-VA, (e) Post-resection of arteriovenous malformation, (d) digital subtraction angiography post-resection|
Click here to view
Giant arteriovenous malformation (Spetzler Martin grade IV and V arteriovenous malformations)
On the basis of their size alone, giant AVMs are more likely to have a significant component of central venous drainage and have a portion of the AVM adjacent to or within the ventricle. Radiosurgery can be used as a preoperative adjunct to obliterate portions of giant AVMs, thus decreasing the size of the remaining nidus for later surgical resection; many giant AVMs can be treated safely with the use of currently available technology., Our case (S2 V1 E1) was managed similarly with preoperative embolisation, coiling and staged gamma knife radiosurgery and the other case was managed with preoperative embolisation, coiling and microsurgery, but still our both patients had poor outcome (mRS > 2).
| Conclusion|| |
Microsurgical management of cerebral AVM is necessary for SMG I and II, with accurate preoperative embolisation aided by optimal selection of the surgical approach and meticulous haemostasis by the coagulation of the deep small difficult vessels (inside apparently normal brain around the AVM) by using surgical adjuncts such as intra operative ICG and FLOW 800, and with adequate follow up of the patients both clinically and radiologically. In our study, microsurgical management of brain AVM has better outcome than endovascular management. Younger age and SMG I and II are the predicting factors for good outcome.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Buil PR, Rubio NS, Quintana EM, Vazquez AB, VELASCO A, Vazquez AA, et al
. Brain Arteriovenous Malformations: Imaging findings and treatment aproaches. European Congress of Radiology-ECR 2013.
Article R. Predicament and challenges in the treatment of intracranial arteriovenous malformations The epidemiology of AVMs Clinical manifestation of AVMs Treatment of AVMs. 2017;1:8-12.
Shankar JJS, Menezes RJ, Pohlmann-Eden B, Wallace C, TerBrugge K, Krings T. Angioarchitecture of brain AVM determines the presentation with seizures: Proposed scoring system. Am J Neuroradiol 2013;34:1028-34.
Sackey FN, Pinsker NR, Baako BN. Highlights on cerebral arteriovenous malformation treatment using combined embolization and stereotactic radiosurgery: why outcomes are controversial?. Cureus. 2017;9.
Inoue HK. Long-term results of Gamma Knife surgery for arteriovenous malformations: 10- to 15-year follow up in patients treated with lower doses. J Neurosurg [Internet]. 2006;105 Suppl:64-8. Available from: http://www.ncbi.nlm.nih.gov/pubmed/18503332
. [Last accessed on 2018 Sep 28].
Lawton MT, Solomon RA, Raso J, De Oliveira E, Han PP, Spetzler RF. Spetzler-Martin Grade III arteriovenous malformations: Surgical results and a modification of the grading scale. Neurosurgery 2003;52:740-9.
Lawton MT. The Role of AVM Microsurgery in the. 2015;4:2014-5.
Wong J, Slomovic A, Ibrahim G, Radovanovic I, Tymianski M. Microsurgery for ARUBA trial (A Randomized Trial of Unruptured Brain Arteriovenous Malformation)–eligible unruptured brain arteriovenous malformations. Stroke 2017;48:136-44.
Killory BD, Nakaji P, Gonzales LF, Ponce FA, Wait SD, Spetzler RF. Prospective evaluation of surgical microscope-integrated intraoperative near-infrared indocyanine green angiography during cerebral arteriovenous malformation surgery. Neurosurgery 2009;65:456-62.
Roessler K, Krawagna M, Dörfler A, Buchfelder M, Ganslandt O. Essentials in intraoperative indocyanine green videoangiography assessment for intracranial aneurysm surgery: conclusions from 295 consecutively clipped aneurysms and review of the literature. Neurosurgical focus. 2014;36:E7.
Alander JT, Kaartinen I, Laakso A, Pätilä T, Spillmann T, Tuchin VV, Venermo M, Välisuo P. A review of indocyanine green fluorescent imaging in surgery. Int J Biomed Imaging 2012;2012.
De Gasperi A, Mazza E, Prosperi M. Indocyanine green kinetics to assess liver function: Ready for a clinical dynamic assessment in major liver surgery? World J Hepatol. 2016;8:355-67.
Lopez KA, Waziri AE, Granville R, Kim GH, Meyers PM, Connolly Jr ES, et al
. Clinical usefulness and safety of routine intraoperative angiography for patients and personnel. Neurosurgery. 2007;61:724-30.
Sato T, Suzuki K, Sakuma J, Takatsu N, Kojima Y, Sugano T, et al
. Development of a new high-resolution intraoperative imaging system (dual-image videoangiography, DIVA) to simultaneously visualize light and near-infrared fluorescence images of indocyanine green angiography. Acta neurochirurgica. 2015;157:1295-301.
Lawton MT, Du R, Tran MN, Achrol AS, McCulloch CE, Johnston SC, et al
. Effect of presenting hemorrhage on outcome after microsurgical resection of brain arteriovenous malformations. Neurosurgery. 2005;56:485-93.
Feng S, Zhang Y, Sun Z, Wu C, Xue Z, Ma Y, et al
. Application of Multimodal Navigation together with Fluorescein Angiography in Microsurgical Treatment of Cerebral Arteriovenous Malformations. Sci Rep [Internet]. 2017;2017:1-9.
Larsen AMG, Cote DJ, Dubois HM, Smith TR. National Databases for Neurosurgical Outcomes. 2018;83:333-44.
Beecher JS, Lyon K, Ban VS, Vance A, McDougall CM, Whitworth LA, et al
. Delayed treatment of ruptured brain AVMs: is it ok to wait? J Neurosurg [Internet]. 2017;128:1-7.
Nataraj A, Mohamed MB, Gholkar A, Vivar R, Watkins L, Aspoas R, et al
. Multimodality Treatment of Cerebral Arteriovenous Malformations. World Neurosurg [Internet]. 2014;82:149-59.
Ellis JA, Lavine SD. Role of embolization for cerebral arteriovenous malformations 2014;(4):234-9.
Turjman F, Massoud TF, Sayre JW, Viñuela F, Guglielmi G, Duckwiler GR. Epilepsy associated with cerebral arteriovenous malformations: A multivariate analysis of angioarchitectural characteristics. undefined [Internet]. 1995 [cited 2018 Sep 23];
Lawton MT, Brain U, Study AVM. malformations. 2016;122:912-20.
Lees R. The pattern of cognitive impairment in stroke survivors with carotid stenosis. 2014.
Cognard C. A randomized trial of unruptured brain arteriovenous malformations study: What impact on clinical care and therapeutic decision? Am J Neuroradiol. 2015;36:619–22.
Morgan MK, Rochford AM, Tsahtsarlis A, Little N, Faulder KC, Da Silva RT, et al
. Surgical Risks Associated with the Management of Grade I and II Brain Arteriovenous Malformations. Neurosurgery. 2004;54:832-9.
Ren Q, He M, Zeng Y, Liu Z, Liu H, Xu J. Microsurgery for intracranial arteriovenous malformation: Long-term outcomes in 445 patients. PLoS One. 2017;12:1-13.
Chang SD, Marcellus ML, Marks MP, Levy RP, Do HM, Steinberg GK, et al.
Multimodality treatment of giant intracranial arteriovenous malformations. Neurosurgery. 2003;53:1-13.
England TN. Numb Er 24 of Surgical Intervention. 1998;1725-33.
[Figure 1], [Figure 2]