Year : 2006 | Volume
: 16 | Issue : 4 | Page : 757--764
The brain arterio-venous malformations (BAVMs): A pictorial essay with emphasis on role of imaging in management
S Kumar, AM Patel, DU Vaghela, K Singh, RN Solanki, HR Shah
Department of Radiodiagnosis & Imaging, B. J. Medical College, Civil Hospital, Ahmedabad, India
42, Nirant Park, Part-1, Opp. Sun-n-Step Club, Thaltej, Post-Ghatlodia, Ahmedabad-380061
|How to cite this article:|
Kumar S, Patel A M, Vaghela D U, Singh K, Solanki R N, Shah H R. The brain arterio-venous malformations (BAVMs): A pictorial essay with emphasis on role of imaging in management.Indian J Radiol Imaging 2006;16:757-764
|How to cite this URL:|
Kumar S, Patel A M, Vaghela D U, Singh K, Solanki R N, Shah H R. The brain arterio-venous malformations (BAVMs): A pictorial essay with emphasis on role of imaging in management. Indian J Radiol Imaging [serial online] 2006 [cited 2020 Jan 29 ];16:757-764
Available from: http://www.ijri.org/text.asp?2006/16/4/757/32341
Arteriovenous malformations are a complex tangle of abnormal arteries and veins linked by one or more fistulas and lacking a capillary bed; the small arteries being deficient in muscularis layer  (Fig 1). They are thought to arise from developmental derangements at embryonic stage of vessel formation at fetal stage or after birth . This pictorial essay is the outcome of review of 77 cases we came across in the last four years. BAVMs account for approximately 11% of cerebrovascular malformations and are more likely than other malformations to be clinically symptomatic. AVMs are important cause of hemorrhage in young adults and account for 2% of all strokes.
BAVMs are classified according to their blood supply:
" Pial or parenchymal (Fig 2A) lie within brain parenchyma and are supplied by the internal carotid or vertebral circulation.
" Dural AVMs (Fig 2B) are uncommon, almost always infratentorial and theorized to be secondary to trauma, surgery, thrombosis of an adjacent venous sinus, or veno-occlusive disease. They receive blood from external carotid circulation and drain into the transverse and sigmoid sinuses, but they may also involve the cavernous sinus, inferior petrosal sinus, superior sagittal sinus or other areas of the venous system.
" Mixed AVMs (Fig 2C) usually occur when it recruits blood vessels from both the internal and external carotid arteries.
Meningeal blood supply (Fig 2D) can be recruited in cases of subtentorial AVM, moderate AVM with steal phenomenon, a large AVM, diffuse AVM and in older patients. By selective external angiography, Newton and Cronqvist have detected meningeal arterial contributions in 48 % cases, including the pure dural type .
BAVMs do not include pure vein of Galen AVMs, cavernous malformations, dural arteriovenous fistulas (DAVF), venous malformations, venous varices or any of the other rarer types of cerebrovascular anomalies .
Of BAVMs, 82% are lobar and 12% are infratentorial while 9% of all are deep (Fig 3). 48% tend to occur at watershed areas (straddling >1 vascular territory) . 70% of supratentorial AVMs are purely pial while approximately 50% of posterior fossa AVMs are purely pial, the rest being purely dural or mixed pial-dural.
BAVMs may be microscopic or large enough to involve an entire hemisphere (Fig 4). BAVM volume has been classified  by the size of the nidus, into four groups: huge (> 50 ml), large (25-50 ml), moderate (10-24 ml), small (<10 ml). Diameters are measured in the middle to late arterial phase of the angiogram, with correction for magnification. Most AVMs are small (<3 cm, 87%) . Angiographically invisible AVMs are termed cryptic vascular malformations, and these possibly representing completely thrombosed AVMs. Multiple BAVMs (Fig 5) can be seen in syndromes like Osler-Weber-Rendu syndrome, Wyburn syndrome etc.
Clinical aspects: Pial AVMs affect both sexes equally. The anterior cranial fossa dural AVMs occur more in men while other dural AVMs are commoner in women. The pial lesions present with hemorrhage, neurological deficits, headache etc. Dural AVMs may show obstruction and thrombosis of sinuses, while anterior cranial fossa dural AVMs may also present with hemorrhage from a ruptured venous aneurysm. Hemorrhage will produce mass effect which otherwise is typically not associated with AVMs.
Factors increasing risk of a first hemorrhage are: small AVM, exclusively deep venous drainage, high intranidal pressure reflected by high pressure in feeding artery and restriction of venous outflow . Doung DH et al have found that high arterial input pressure and venous outflow restriction are more important than size, location and associated aneurysms .
Spetzler and Martin grading system  (Table 1): The grade of a lesion is determined by summing the points in each of the 3 categories. Grades I and II have low morbidity. Surgical treatment of a grade I, II and III AVM presents little risk of morbidity and mortality. By contrast, a grade IV lesion is associated with 31% significant risk while Grade V lesion has 50% treatment associated morbidity. A grade VI AVM is considered inoperable.
BAVMs may be either compact (66%) or diffuse (34%) . Compact AVMs (Fig 6) have a nidus formed by tightly packed entangled venous loops interconnected by small venules. When located supratentorially, compact AVMs are often wedge shaped extending through both gray matter and white matter with the base parallel to the meninges.
Feeding artery of BAVM is defined as any intracranial vessel that angiographically contributes arterial flow to the malformation. Feeding arteries may be parent arteries that give rise to vessels that directly or indirectly supply flow to the AVM. Coding of multiple vessels is also possible . Feeding arteries are of 3 types.
" The circumferential feeding artery (Fig 7) extends around the nidus and sends branches both to small arterioles connected to the nidus and to normal brain capillaries.
" Penetrating feeding arteries (Fig 8) bisect the AVM core and send branches to it.
" Final feeding arteries (Fig 9) either connect directly to an AVM loop or branch to shunting arterioles.
The venous drainage pattern is categorized as superficial, deep or both  (Fig 10 A, B, C, D, E). Associated venous varix is seen in 16% cases  (Fig 11 A, B).
BAVMs are hemodynamically compartmentalized; each compartment has its own feeding arteries and draining veins. The number of compartments in an AVM is proportional to its size. An AVM smaller than 3 cm in diameter is likely to have 1 compartment, a 3-cm or 4-cm AVM may have 2 compartments, and an AVM larger than 4 cm in diameter typically has at least 3 compartments. Blood flow through an AVM is proportional to the number of compartments and to AVM volume.
Diffuse AVMs (Fig 12) are dispersed among normal brain tissue and are typically found in the basal ganglia or thalamus.
Regional blood flow surrounding an AVM may be reduced to 81% of normal referred as the steal phenomenon (Fig 13) although the total cerebral blood flow may be increased by as much as 50-100%.
Due to ischemia, the neighboring parenchyma undergoes atrophy and gliosis (Fig 14) and discolored by hemosiderin after prior hemorrhage, with production of scattered foci of calcification (Fig 15), due to the slow progressive growth.
Concurrent arterial aneurysms (Fig 16 A, B) can occur in association with BAVMs in 10-58% of patients  and are defined as saccular dilatations of the lumen 2 times the width of the arterial vessel that carries the dilatation and are further classified as feeding artery aneurysms, intranidal aneurysms, and aneurysms unrelated to blood flow to the AVM. Intranidal aneurysms are coded only when visualized early after angiographic injection, e.g, before substantial venous filling occurs. Infundibula, arterial ectasias (i.e., dilated feeding vessels), and intranidal aneurysmal dilatations seen during the venous angiographic phase only are not coded as arterial aneurysms .
CTA has proven effective because it provides the simultaneous visualization of arterial and venous anatomy including the nidus . CTA is a useful diagnostic technique, both during stereotactic localization before surgical resection or radiosurgical treatment and during the imaging follow-up after radiosurgery .
MRI can demonstrate areas of AVM involvement with its size and shows both the dilated feeding arteries and enlarged draining veins. The associated aneurysms on arterial feeders and associated sequelae such as mass effect, edema, or ischemic changes are also discernible. BAVMs difficult to see on CT due to large hemorrhage can be evaluated by MR (Fig 17). MRI is particularly well suited to document AVM rupture. Unfortunately, the sensitivity of MRI to aneurysms smaller than 1-2 cm is low. MRI is an excellent preoperative planning tool for delineating the relationship between an AVM nidus and critical brain structures, particularly with functional MRI. Certain lesions hidden on conventional angiograms may be identified only on MRIs because of their ability to depict hemosiderin deposits or other evidence of blood breakdown.
MRA is a noninvasive alternative to conventional angiography. 3D dynamic MRI with multiple surface coils and parallel images is the best used yet with least interobserver discrepancy and depiction of direction, rate, and quantity of blood flow which is especially important if embolotherapy is planned. An advantage over DSA is its usefulness in arterial hypertension and clotting disorders. TOF MRA fails to show nidus with slow flow. This pictorial essay shows phase contrast MRA images taken at 30 and 50 cm/s velocity settings.
Postoperative MRI is useful to study the effect of surgery on the adjacent brain and will show the extent of nidal, arterial, or venous thrombosis following embolisation; however, documentation of complete obliteration of the nidus is performed best with conventional angiography because MRI may fail to depict small amounts of residual nidus or persistent AV shunting and because the material used for embolisation interferes on MRA.
Catheter angiography is the best modality to delineate the architecture. It can be used to measure the size of the AVM (Fig 18), judge the compactness of the nidus and to evaluate the venous drainage pattern along with depiction of aneurysms and venous stenosis. Planning angiography remains vital in both interventional neuroradiologic and neurosurgical evaluation of patients with AVM. The goal of the study should be to identify the number and location of feeding arteries, the angiographic location and size of the nidus, the shunt type of the lesion (e.g., high flow vs. low flow) pertaining to both ICA and ECA contributions and the pattern of venous drainage (e.g., superficial, deep or mixed).
Houdart E et al have proposed an angiographic classification of intracranial arteriovenous fistula and malformations  (Fig 19). Arteriovenous fistulae have maximum three arterial feeders reaching a single vein. A single arteriovenous channel is an example. Arteriolovenous fistulae have a plexiform arterial structure with several arteries feeding shunts in the wall of a single draining vein. Arteriolovenulous fistulae also have plexiform structure but the shunts have symmetrical arrangements with each afferent arteriole facing its efferent venule and the initial venous compartment being away from the shunts. Pial AVMs are generally arteriolovenulous type while dural AVMs are arteriolovenous type. Transarterial or transvenous embolisation can be done in arteriovenous and arteriolovenous fistulas while only arterial embolisation is permissible in arteriolovenulous fistula as blocking the vein which is located distally will raise intranidal pressure resulting in hemorrhage.
The treatment of an intracranial AVM typically involves embolisation (Fig 20 A, B), direct surgical or microsurgical resection or radiosurgery. Usually, small or medium-sized AVMs located in noncritical areas of the brain can be removed successfully with conventional microsurgery. In contrast, large BAVMs or AVMs in eloquent cerebral locations usually require staged, multimodal treatment. Embolisation is performed alone when surgery is inadvisable or refused by the patient or can be done prior to surgery to reduce the volume of the AVM nidus or may follow microsurgery, radiosurgery, or both. Many different agents have been used for embolisation ; however, glue embolization with cyanoacrylates (NBCA) allows a permanent and complete cure of AVM. The complications resulting from embolization include hemorrhage, retrograde thrombosis, and inadvertent occlusion of nontarget vessels, normal perfusion pressure breakthrough (NPPB) and thrombosis of vein which drains both the AVM and normal adjacent brain.
|1||The Arteriovenous malformation group. Arteriovenous malformation of the brain in adults. NEJM June 1999; no.23, vol 340:1812 - 1818.|
|2||Newton TH, Cronqvist S (1969) Involvement of dural arteries in intracranial arteriovenous malformations. Radiology 93: 1071- 1078.|
|3||Joint Writing Group of the Technology Assessment Committee American Society of Interventional and Therapeutic Neuroradiology. Reporting Terminology for Brain Arteriovenous Malformation Clinical and Radiographic Features for Use in Clinical Trials. Stroke. 2001; 32:1430.|
|4||C. Stapf, A.V. Khaw, R.R. Sciacca, et al. Effect of Age on Clinical and Morphological Characteristics in Patients with Brain Arteriovenous Malformation. Stroke. 2003; 34:2664.|
|5||Pasqualin A, Barone G, Cioffi F et al. The relevance of anatomic and hemodynamic factors to a classification of cerebral artefiovenous malformations. Neurosurgery 1991; 28:370-379.|
|6||Bruce E. Pollock, John C. Flickinger, L. Dade Lunsford et al. Factors That Predict the Bleeding Risk of Cerebral Arteriovenous Malformations. Stroke. 1996; 27:1-6.|
|7||Spetzler RF, Hargraves RW, McCormack PW et al. Relationship of perfusion pressure and size to risk of hemorrhage from AVMs. J. Neurosurgery 1992; 76:918 -923.|
|8||Duong DH et al. Feeding artery pressure and venous drainage pattern are primary determinants of hemorrhage from cerebral arterivenous malformations. Stroke 1998; 29:1167 -1176.|
|9||Spetzler RF, Martin NA. A proposed grading system for arteriovenous. J Neurosurgery 1986; 65: 476-483.|
|10||C. Stapf, H. Mast, R.R. Sciacca et al for the New York Islands AVM Study Collaborators. The New York Islands AVM Study Design, Study Progress, and Initial Results. Stroke. 2003; 34:e2. |
|11||Redekop G, TerBrugge K et al .Arterial aneurysm associated with cerebral arteriovenous malformation: classification, incidence and risk of hemorrhage. J Neurosurgery1998; 89:539-546.|
|12||J. Rieger,N.Hosten, K. Neumann et al. Initial clinical experience withspiral CT and 3D arterial reconstruction in intracranial aneurysms and arteriovenous malformations. Neuroradiology 1996; 38:245- 251. |
|13||Jean-Yves Gauvrit, Catherine Oppenheim, Francois Nataf et al. Three-dimensional dynamic magnetic resonance angiography for the evaluation of radiosurgically treated cerebral arteriovenous malformations. Eur Radiol 2006; 16: 583-591.|
|14||Houdart E, Gobin YP, Casaco A et al. A proposed angiographic classification of intracranial arteriovenous fistulae and malformations.Neuroradiology 1993, 35:381-385.|
|15||Oglvy CS et al. Recommendations for management of intracranial arteriovenous malformation. Stroke2001; 32:1458.|