Endovascular Treatment of Cerebral Arteriovenous Malformations

Significant advancements in the management and treatment of cerebral arteriovenous malformations (AVMs) have been made over the last few years. Although surgical resection remains the most definitive treatment option in selected patients, both endovascular and radiosurgical techniques have been added to the arsenal available for the treatment of these lesions. These treatment modalities can be used individually or combined in a multimodality approach. This chapter takes a closer look at endovascular techniques in the treatment of AVMs.

Pretherapeutic Evaluation

Noninvasive Techniques

Cerebral AVMs can be visualized with a variety of diagnostic modalities, including computed tomography (CT), computed tomography angiography (CTA), magnetic resonance imaging (MRI), and magnetic resonance angiography (MRA). Noncontrast CT scans have a low sensitivity but enable evaluation of acute hemorrhage. Prominent draining veins are often seen on noncontrast scans. CTA provides important vascular detail within the AVM, including the enlargement of feeding arteries, engorged cortical veins, or the presence of associated aneurysms. However, it provides lower definition of surrounding cerebral structures, which MRI and MRA can show better. MRI characterizes AVMs by exhibiting flow voids on T1 and T2 imaging, with occasional hemosiderin deposits suggesting prior hemorrhage. Additional modalities have been used in the preoperative evaluation of AVMs in efforts to understand the anatomic as well as functional relationship of the brain to the vascular malformation.

Functional Magnetic Resonance Imaging

Functional MRI quantifies the local increase in oxyhemoglobin concentration secondary to increased blood flow and volume that occurs after stimulation of the eloquent cortex. The images enable the clinician to recognize the relationship of AVMs to eloquent brain structures as well as to elucidate cortical reorganization secondary to the AVM. This information may help during preoperative planning and in patient counseling because it can, in some measure, predict the development of posttherapy deficits.

Diffusion-Tensor Imaging

Diffusion-tensor imaging is a developing technology that enables in vivo visualization of multiple white matter tracts and their relationship to an AVM. This information can be used to help minimize treatment risks and determine the best therapeutic modality for each patient. ^,

Invasive Techniques

Provocative Injection Testing (Wada Test)

Prior to embolization of an AVM, Wada testing of selective feeding vessels can be performed. Amobarbital sodium or propofol injections have been performed on vessels in or remote from the lesion. This pharmacologic test helps to predict the effect of occluding a feeding vessel. Due to the short half-life of these injections, their effect disappears rapidly. Such evaluation helps prevent embolization of pedicles that may result in focal neurologic deficit. This preoperative functional evaluation is helpful when supraselective intranidal microcatheter placement is not achieved and there is concern that vessels to adjacent normal brain may be compromised.

Angiography

Angiography remains the gold standard for defining the arterial and venous anatomy of an AVM. Such analysis of the angioarchitecture helps in evaluating the risk of hemorrhage and selecting the best therapeutic management of the AVM. Angiographic characterization of the AVM includes identification of the main arterial feeders, size of the AVM, shape of the nidus (compact vs. diffuse, without clear borders that separate it from adjacent brain), drainage patterns (superficial drainage into cortical veins, deep drainage through the deep venous system, or proximity to dural sinuses), and characteristics of outflow (presence of restriction from stenosis or sinus thrombosis). Anatomic characteristics of AVMs such as deep location, deep venous drainage pattern, presence of a single draining vein, venous stenosis, eloquent location, and small diameter are all related to increased risk of hemorrhage or neurologic deficit.

Angiography helps to elucidate the presence of associated aneurysms. These are categorized by their location (i.e., feeding artery, intranidal, circle of Willis, or venous). Associated aneurysms are found in approximately 7% to 41% of patients with AVMs. Studies evaluating hemorrhage risks for AVMs with associated aneurysms have shown conflicting results. A 2007 study from the Columbia study group described prospectively collected hemorrhage risk factors limited to the period between the initial AVM diagnosis and the start of treatment, calculating annual hemorrhage rates in 622 patients. In the group’s model, the presence of associated intranidal or feeding artery aneurysms did not have a significant effect on hemorrhage rates. Other studies, however, have included intranidal aneurysms as important risk factors for hemorrhage in AVMs, ^, ^, ^, ^, with increased size of the aneurysm related to increased risk of hemorrhage. ^, These aneurysms are exposed to the same intraluminal pressures as the arterial components of the AVM. Therefore any pressure changes following AVM embolization can lead to associated aneurysmal rupture. Thus vessels that supply intranidal aneurysms should be recognized and treated early to reduce the risk of hemorrhage.

Angiography further evaluates for the presence of pseudoaneurysms (false aneurysms), which may form in previously ruptured AVMs. These irregularly shaped cavities have abnormally structured vessel walls and are often located at the point where the artery and vein make contact due to inherent weakness. Previous rupture of an AVM is an independent predictor of future hemorrhage, and recognizing the presence of posthemorrhage pseudoaneurysm formation can help reduce the risk of future hemorrhage.

AVM-associated arteriovenous fistulas are sometimes discovered in the nidus during angiographic evaluation. The presence of these high-flow arteriovenous fistulas (AVFs) is associated with increased risk of intra- and postoperative complications, including hemorrhage. ^, Prior endovascular treatment of these AVFs helps to decrease the chance of this complication. Reducing the flow through the AVF also helps to decrease the chances of undesired embolization in the AVMs’ venous drainage as well as in the cerebral veins, dural sinuses, or pulmonary circulation. ^,

Timing of imaging after a hemorrhage must be considered. Performance of an angiogram immediately after hemorrhage can produce a falsely negative result secondary to compression of the nidus by the hematoma. Therefore a late angiogram, performed 3 months after the hemorrhage, remains the gold standard for the detection of an AVM.

Endovascular Technique

Anesthetic and Perioperative Considerations

The type of anesthesia is selected depending on the angioarchitecture of the AVM. If the target vessel is terminal, ending directly in the AVM, the procedure can be done under general anesthesia. This facilitates patient comfort and ensures a motionless patient during the procedure, enabling improved visualization of the AVM structures and minimizing patient risk associated with mobility during the procedure. Because there is no functional intervening tissue within the AVM, embolization of appropriate vessels should theoretically not cause any functional deficit. In our institution, general anesthesia and endotracheal intubation are utilized for all AVM embolizations. A brief period of apnea helps decrease the motion caused by the respiratory cycle in certain cases when rapid injection of embolic material and removal of a microcatheter is necessary. Intraoperatively, the patient’s neurologic status is monitored continuously via somatosensory evoked potentials and electroencephalography.

In some cases it is necessary to assess the patient’s functional status with provocative tests intraoperatively. This requires intravenous anesthesia with short-acting agents such as propofol and midazolam and avoiding paralytic agents so that the patient’s neurologic function can be assessed quickly during the procedure. Authors that favor this approach point to the wide variability and cortical reorganization described in patients with AVMs. ^,

Foley catheters and continuous transduction of arterial pressure are standard. Blood pressure transduction can be obtained through the femoral sheath or existing arterial line. This is especially important when vasoactive drugs must be given intraprocedurally. Deliberate systemic hypotension through vasoactive agents, general anesthesia, or adenosine-induced cardiac pause may also be used to slow flow through the AVM and allow for more controlled deposition of embolic material. In addition, a pulse oximeter is applied to the foot ipsilateral to the femoral sheath to monitor signs of early vessel obstruction, distal thromboemboli, or overcompression following sheath removal. Foley catheters are typically used for fluid management, and supplemental oxygen or nasopharyngeal airways can be considered, especially in those patients under a sedative-hypnotic agent.

Standard Procedure

The goals, expectations, risks, and benefits of endovascular treatment must be discussed with the patient and family prior to the procedure. Vascular access is obtained by inserting a 7-French sheath into the femoral artery via a Seldinger technique. Anticoagulation algorithms to prevent thromboembolic complications during and after the procedure remain controversial. In our institution, a continuous heparinized flush is used for the femoral sheath, guide catheter, and microcatheter. However, no additional heparin bolus is used during the procedure. A 6-French guiding catheter is advanced into the internal carotid or vertebral artery. Flow-directed microcatheters—their selection based on the size of the vessel—are used to reach the intranidal target. The catheter, which contains a flexible distal tip to enable navigation through distant and tortuous neurovasculature, is drawn through the vessels primarily via blood flow. Some flow-directed catheters have a steam-formed distal tip to enable more selective tracking of the catheter into the desired vessels. Variations in microcatheters—including those in Prowler (Cordis Endovascular, Miami Lakes, FL), Spinnaker (Boston Scientific/Target Therapeutics, Fremont, CA), Marathon (ev3 Neurovascular, Irvine, CA), and Echelon (ev3 Neurovascular)—allow for differences in flexibility, torque, maneuverability, and responsiveness.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here