Vertebral Artery Interventions: A Comprehensive Updated Review

Abstract

Patients with posterior circulation ischemia due to vertebral artery stenosis account for 20 to 25% of ischemic strokes and have an increased risk of recurrent stroke. In patients treated with medical therapy alone, the risk of recurrence is particularly increased in the first few weeks after symptoms occur, with an annual stroke rate of 10 to 15%. Additionally, obstructive disease of the vertebrobasilar system carries a worse prognosis, with a 30% mortality at 2-years if managed medically without additional surgical or endovascular intervention.

Percutaneous transluminal angioplasty and stenting of symptomatic vertebral artery stenosis are promising options widely used in clinical practice with good technical results; however, the improved clinical outcome has been examined in various clinical trials without a sufficient sample size to conclusively determine whether stenting is better than medical therapy. Surgical revascularization is an alternative approach for the treatment of symptomatic vertebral artery stenosis that carries a 10-20% mortality rate.

Despite the advances in medical therapy and endovascular and surgical options, symptomatic vertebral artery stenosis continues to impose a high risk of stroke recurrence with associated high morbidity and mortality. This review aims to provide a focused update on the percutaneous treatment of vertebral artery stenosis, its appropriate diagnostic approach, and advances in medical therapies.

1. INTRODUCTION

Stroke is the fifth most prevalent cause of death and a major cause of disability for adults in the United States. Ischemic strokes account for 87% of all strokes, with 10% due to intracranial hemorrhage (ICH) and 3% due to subarachnoid hemorrhage (SAH) [1]. Furthermore, strokes account for a large proportion of death and disability worldwide, responsible for an estimated global disease burden of about 51 million disability-adjusted life years among patients aged 20-64 years [2].

Posterior circulation ischemic stroke has an estimated prevalence of 20% to 25% [3]. Vertebral arterial occlusion in patients with known atherosclerotic peripheral artery disease (PAD) has a 40% incidence of vertebral artery stenosis (VAS). In the New England Medical Center Posterior Circulation Registry, 82 of 407 patients with ischemia affecting the posterior circulation had >50% stenosis of the extracranial vertebral artery. Annual stroke rates for patients with symptomatic intracranial vertebral and basilar artery stenosis are 8% and 11%, respectively. There is an increased 90-day risk of recurrent vertebrobasilar stroke in patients with at least 50% atherosclerotic stenosis of the vertebrobasilar artery. Treatment of vertebrobasilar insufficiency with medical therapy alone is associated with an annual stroke rate of 10% to 15% [4, 5]. Additionally, the prognosis of thrombotic or atherosclerotic occlusion of the vertebrobasilar arterial system is poor, with a 30% mortality at 2-years for medically treated strokes without additional surgical or endovascular intervention [5].

Atherosclerotic disease of the vertebrobasilar system may affect any segment of the vertebral artery (V.A.) but is more common in the ostium and proximal segment. Symptomatic VAS is managed either medically or via endovascular or surgically revascularization. The morbidity and mortality of endovascular compared to surgical revascularization of both intracranial and extracranial VAS have been reported to be reduced in the literature [6-8]. Despite the advances in medical therapy and endovascular and surgical options for the treatment of VAS, symptomatic VAS continues to impose a high risk of stroke recurrence with associated high morbidity and mortality.

2. ANATOMY OF THE VERTEBRAL ARTERY

The vertebral artery arises from the superior-posterior wall of the subclavian artery and is divided into four segments (V1 to V4). The initial segment (V1) extends from its origin at the subclavian artery to the transverse foramina of the 5^th or 6^th cervical vertebra. The V1 segment can be treated percutaneously with relative ease when it is straight and not tortuous or redundant, as it tends to be in the elderly population [9]. The V2 segment passes through the bony canal of the transverse foramina from C2 to C6. Percutaneous treatment is favorable in this segment due to its short distance to the subclavian artery and straight course. The V3 segment originates at the transverse foramina of C2 and terminates at the dura mater. This segment is particularly tortuous to allow mobility of the atlantoaxial and atlantooccipital joint; however, it increases the complexity of the potential percutaneous intervention, and therefore, balloon-expandable stents should be avoided in this region. Avoiding intervention of extreme tortuous arteries and using short self-expanding stents may offer treatment success in this segment [5]. The V4 segment extends intracranially along the inferior portion of the pons and joins the contralateral V.A. to form the basilar artery. The anterior spinal communicator arteries originate from this segment, join in the midline, and form the anterior spinal artery. The intervention of this segment should be performed with extreme caution since it supplies the anterior two-thirds of the spinal cord, and occlusion of the anterior spinal communicator arteries can cause significant deficits [5, 9]. Segments of the V.A. are illustrated in Fig. (1).

Fig. (1).

Vertebral artery anatomy. (From Jenkins JS, Collins TJ. Vertebrobasilar insufficiency. In: Jaff MR, White CJ, editors. Textbook of vascular disease, diagnostic and therapeutic approaches. Minneapolis (M.N.): Cardiotext Publishing, LLC; 2011. p. 99).

Anatomic variants of V.A. are much more common than variants of the carotid circulation. The vertebral arteries typically arise from the subclavian arteries; in approximately 5% of individuals, however, the left V.A. arises directly from the aortic arch. The diameter of the left V.A. diameter is larger than (in 50% of individuals) or equal to (in 25% of individuals) that of the right vertebral artery. In approximately 10% of people, 1 V.A. is markedly smaller than the other. When this is the case, the smaller V.A. may terminate in the posterior inferior cerebellar artery or have a hypoplastic segment that extends beyond the posterior inferior cerebellar artery to the basilar artery, contributing little to the basilar artery blood flow. These important anatomic variations must be considered in clinical assessment and treatment.

3. MECHANISM OF VERTEBRAL ARTERY STENOSIS

Vertebral artery stenosis can occur via either a major/common mechanism or a minor/less common mechanism. The major mechanism of VAS includes atherosclerosis, while the minor/less common mechanism includes congenital disease, arterial dissection, trauma, or extrinsic compression, and vasculitis (such as Takayasu diseases or Giant cell Arthritis).

As with most other arteries, atherosclerosis is a major mechanism of VAS, and the process is the same. It starts with an intimal accumulation of lipoprotein particles which then undergo oxidative modification with the elaboration of cytokines, resulting in the expression of adhesion molecules and chemoattractants that facilitate uptake and migration of monocytes into the artery wall. Lipid-laden macrophages or form cells are formed from these monocytes, accumulating additionally modified lipoproteins and releasing additional cytokines, oxidants, and matrix metalloproteinases. Migration, proliferation, and elaboration of smooth muscle cells in the intima with an accumulation of the extracellular matrix with the connective tissues form the fibrous cap. Initially, atherosclerotic lesion grows outward direction in the process of arterial remodeling, but as it continues to grow, it encroaches the lumen, resulting in a reduction in the caliber of the artery and further narrowing, thereby resulting in severe stenosis by plaques disruption and formation of thrombus. Plaque expansion can also occur when there is a rupture of friable micro-vessels at the base of the plaque, which results in intraplaque hemorrhage. Plaque disruption can also occur, including fibrous cap, superficial erosion, and erosion of a calcium nodule.

It is important to note that in the vertebral artery, the proximal portion or the origin of the V.A. is prone to atherosclerotic plaques and subsequent stenosis, but atheromatous plaques in the V.A. have been shown to be smoother and not prone to ulceration, as seen in the coronary artery.

Congenital vertebral diseases, such as congenital arterial hypoplasia and atresia, can result in VAS; the commonest congenital vertebral disease is V.A. hypoplasia, with a reported frequency of 2-6% from autopsy and angiogram. V.A. hypoplasia has been associated with the symptomatic vertebrobasilar occlusive disease.

Extrinsic compression of the V.A. resulting in stenosis can be due to trauma to the cervical vertebra, usually affecting the second or the third part of the cervical vertebra, or osteophytes impingement and compression of the vertebral artery.

Vertebral artery dissection may occur due to blunt trauma that results in the separation of the arterial wall layers. In the case of the vertebral artery, stenosis is usually the subintimal dissection, leading to an intramural hematoma and subsequently resulting in stenosis or occlusion.

In summary, the major mechanism of VAS still plays a significant role in the mechanism, but other minor mechanisms are supposed to be considered when planning for endovascular or surgical repair of V.A.

4. RISK FACTORS AND CLINICAL MANIFESTATION

Risk factors for the stroke of the anterior and posterior circulation for patients with coronary and peripheral artery disease include hypertension, hyperlipidemia, age, gender, tobacco use, family history, genetics as well as cardio-embolic conditions. Patients with posterior circulation stroke may display a wide variety of symptoms of vertebrobasilar ischemia involving different parts of the brain (brainstem, cerebellum, thalamus, occipital visual cortex, medial temporal lobe, and auditory/vestibular structures).

Occlusion of the extracranial V.A. causes ischemia in the medulla and/or cerebellum and usually manifests as brief TIAs. Conversely, occlusion of the intracranial V.A. causes ischemia in the lateral medulla, leading to Wallenberg syndrome (Horner’s syndrome, dysphagia, hoarse voice, limb ataxia, and decreased pain/ temperature sensation of the ipsilateral face and contralateral body) [10]. Basilar artery occlusion can cause pontine damage leading to “Locked-In-Syndrome,” in which patients display quadriplegia and mutism with preserved consciousness [11]. Distal basilar artery occlusion typically occurs from embolic sources, such as cardiac or V.A. sources, resulting in ischemia of the rostral midbrain and thalamus. These patients present with prodromal symptoms of vertigo, nausea, headache, neck pain, and transient lateralized motor weakness preceding the stroke [12]. Additionally, thromboembolic occlusion of the top of the basilar artery, also known as Rostral Brainstem Infarction, may manifest with visual and oculomotor deficits and behavioral abnormalities; however, motor dysfunction is often absent [13].

5. DIAGNOSIS

Imaging is the gold standard diagnostic modality for VAS; however, the importance of history and physical examination cannot be overemphasized. These imaging modalities include Computerized tomography (C.T.), Magnetic Resonance Imaging (MRI), Doppler Ultrasound (DUS), Contrast-Enhanced Magnetic Resonance Angiography (CEMRA), Digital Subtraction Angiography (DSA), and Computerized Tomography Angiography (CTA) [14]. Approximately 90% of VAS occurs at the origin of the vertebral artery. It is easily visualized with ultrasound (U.S.) and, therefore, it is the initial non-invasive test of choice. U.S. alone can visualize the origin of the V.A. in up to 60% of patients, and with the use of Doppler imaging, up to 80% [15]. Furthermore, transcranial Doppler ultrasound can detect intracranial VAS with a sensitivity of 80% and specificity of up to 97%; however, distal segments can be difficult to visualize [16]. Consequently, waveform abnormalities of the distal vertebral artery, a specific finding for the presence of distal stenosis, are generally used to infer the presence of distal VAS [14, 17]. Furthermore, DUS can estimate V.A. size and flow direction, which helps differentiate between hypoplasia, stenosis, occlusion, and aplasia [14, 17, 18].

In a prospective case series by Yurdakul, the Peak Systolic Velocity Ratio (PSVr) was used to accurately detect proximal VAS, and a ratio >2.2 between intra-stenotic and post stenotic V.A. was considered the optimal criterion for diagnosing ≥ 50% proximal V.A. stenosis with sensitivity and specificity of 96% and 89%, respectively. Other parameters used to identify proximal V.A. stenosis of 50% or more include Peak Systolic Velocity (PSV) greater than 108 cm/s, End Diastolic Velocity (EDV) greater than 36 cm/s, and End Diastolic Velocity ratio (EDVr) greater than 1.7 [19].

MRI, in combination with MRA, can detect both extracranial and intracranial vertebral arteries, and it also poses a higher sensitivity in detecting basilar stenosis and small infarcts [13]. Furthermore, helical or spiral C.T. angiography can detect most extracranial vertebral artery stenotic lesions and differentiate between kinked and atherosclerotic stenosed vessels [20]. Although both CTA and contrast-enhanced MRA have a sensitivity of 94% and specificity of 95%, neither can reliably delineate the origin of the vertebral arteries [21-23].

Although angiography is the gold standard for diagnosing atherosclerotic V.A. disease, it is an invasive procedure that carries a 1-2% risk of stroke [22]. A prospective study conducted by Khan et al. compared CE-MRA, CTA, and DUS with intra-arterial angiography and found that CE-MRA and CTA both had the highest sensitivity and specificity for the detection of moderate and severe stenosis. Importantly, CE-MRA had slightly higher sensitivity and specificity compared to CTA [17]. DUS demonstrated low sensitivity and high specificity. This means it will miss many stenoses, but because of its high specificity, if an abnormality is detected on DUS, it is likely to represent stenosis [23]. Vertebral ostial stenoses are overestimated by MRA, while CTA underestimates the degree and prevalence of ostial VAS.

The European Society of Vascular Surgeons 2018 guidelines now recommend Color DUS as the first-line imaging strategy in patients with suspected vertebrobasilar ischemia. However, it must be followed by either CE-MRA or CTA prior to making any decisions on intervention [14].

5.1. Indication for Vertebral Artery Intervention

Vertebral artery revascularization is generally reserved for symptomatic patients with symptoms, such as motor and sensory deficits, visual changes, and cranial nerve deficits, who are refractory to medical therapies. Evaluation of these symptoms should be approached in terms of their aetiologies; embolic, thrombotic, and hemodynamic causes. Various studies have estimated a one-year risk of 5% to 11% incidence of stroke or death in these subsets of patients without intervention; this risk was reduced by 3% with angioplasty in the Stanford series. While there is no clear consensus on the percentage of VAS prompting endovascular or surgical intervention, major randomized control trials, such as the Vertebral Artery Stenting Trial (VAST), which studied the feasibility and safety of vertebral artery stenting in symptomatic patients, and Vertebral Artery Ischemic Trial (VIST), which compared the risk and benefits of vertebral artery angioplasty and stenting in symptomatic patients with best medical treatment, both included patients with ≥50% stenosis in their trials. J. Stephen et al. suggested revascularization in patients with bilateral VAS of ≥70% and patients with unilateral VAS of ≥70% in the presence of occluded or hypoplastic contralateral artery or artery to artery embolism. Reperfusion of chronic total occlusion is contraindicated, while occlusion distal to V4 and basilar artery is discouraged due to potentially harmful complications. The European Society of Cardiology also recommends: “In patients with symptomatic extracranial vertebral stenoses, revascularization may be considered for lesions ≥ 50% in patients with recurrent ischaemic events despite optimal medical management”- Class IIb. It is reasonable to assume that stenosis of ≥50% in one or both vertebral arteries in the setting of relevant symptomatology is an indication for vertebral artery intervention either by endovascular or surgical approach. In addition, therapies are targeted at reducing atherosclerotic risk factors, such as hypertension, hyperlipidaemia, smoking, diabetes mellitus, and other lifestyle modifications are recommended for all patients with V.A. disease. Although high long-term failure rates eventually require surgical intervention, endovascular therapy was found to be effective in treating symptomatic patients with large vessel disease, such as Takayasu’s arteritis, while surgical bypass can decrease the risk of transient ischemic attacks and ischemic strokes when compared with endovascular and medical management in Moyamoya disease. There are no randomized controlled trials comparing the outcomes of vertebral artery interventions using endovascular angioplasty with or without stenting and surgical treatment of VAS; the choice between these two interventions is influenced by factors, such as the anatomic location of the lesion and the level of expertise of the surgical centre. While studies have reported a cumulative 6-year patency rate ≥ 90% with surgical reconstruction, this should be balanced against the high complication rates, such as transient Horner syndrome (21.6%), transient vocal cord palsy (6.1%), stroke (4.1%), and mortality (9.4%).

6. TREATMENT AND PREVENTION

Patients with V.A. stenosis can be asymptomatic or symptomatic. There are no randomized controlled trials (RCT) evaluating the effect of the risk factors or best medical therapy in asymptomatic patients. There is also limited evidence to determine optimal medical therapy for symptomatic patients with posterior circulation ischemia [14, 20]. American Heart Association/American Stroke Association Guideline recommendation for stroke prevention includes risk factor modification, healthy diet, blood pressure management, diabetes control, smoking cessation, and physical activity. Therapeutic regimens include antiplatelet therapy, statin therapy for lipid management, and antihypertensive medications with a goal of blood pressure <140/90 mmHg. For diabetic patients, strict glycemic control is recommended, along with a lower diastolic pressure goal of <85 mmHg [24].

Although there are no trials that have studied an “optimal” medical therapy for recently symptomatic VAS, the Stenting and Aggressive Medical Management for Preventing Recurrent Stroke in Intracranial Stenosis (SAMMPRIS) trial demonstrated that an aggressive medical treatment strategy with dual-antiplatelet, blood pressure control, lipid-lowering therapy with statin medication, glycemic control, and risk factor modification was highly effective for secondary prevention of stroke [25].

Warfarin, on the other hand, in a non-randomized study, did not show a reduction in stroke in the basilar territory or benefit for VAS or posterior circulation disease in general [26]. Nonetheless, anticoagulation is indicated for patients with vertebral artery occlusion secondary to cardiogenic embolism due to atrial fibrillation or mechanical heart valves [27-29].

The development of endovascular therapy with stenting has become an important method for the treatment of VAS. Consequently, clinical trials have highlighted its safety and efficacy [30, 31]. In a prospective study conducted by Jenkins et al. involving 105 patients that underwent stenting of VAS, procedural and clinical success was achieved in 105 (100%) and 95 (90.5%) patients, respectively. The authors concluded that stenting of VAS is a safer and less morbid alternative to open surgery and should be considered first-line therapy for the treatment of V.A. atherosclerotic disease [5].

Despite the technical feasibility of endovascular treatment, there remains limited evidence demonstrating the superiority of endovascular treatment to medical management [31].

7. ENDOVASCULAR TREATMENT APPROACH

7.1. Treatment Considerations

Successful vertebral artery intervention (VAI) requires an understanding of the anatomy of the anterior and posterior circulation in addition to the circle of Willis and its branches. The gold standard in practice includes a comprehensive digital subtraction contrast angiography scan which includes the aortic arch, carotid arteries, and vertebral arteries, which are then correlated with the posterior circulation to determine the culprit vessel and lesion(s) of interest [32].

To accurately define intracranial anatomy in the catheterization laboratory, fluoroscopic digital subtraction techniques are essential. Furthermore, nonionic, iso-osmolar contrast is recommended for intracranial angiography.

The risks, benefits, and alternatives to the procedure should be explained to the patient and allow time for questions in compliance with the provider’s institutional protocols. Care should be taken to explain the risks of VAI, which include death, ischemic or hemorrhagic stroke, access site bleeding, requiring a blood transfusion, paralysis, or potential worsening or lack of improvement of symptoms despite intervention.

All patients should be pre-treated with loading dose aspirin (325 mg) and clopidogrel (300-600 mg) at least 1 day prior to the procedure. If a patient is previously on another P2Y12 inhibitor (ticagrelor, prasugrel), the provider can consider switching to clopidogrel or continuing with their current regimen. Minimal or no sedation should be used during the procedure. Furthermore, to quickly identify complications, continuous neurological monitoring is encouraged. During the VAI, a low-dose weight-adjusted heparin infusion is recommended to achieve an activated clotting time (ACT) greater than 200 seconds. In the setting of heparin-induced thrombocytopenia (HIT), bivalirudin is an acceptable alternative [33].

7.2. Anatomical Considerations

The ostium (Vo) and V1 segments are the most frequently treated percutaneous intervention locations. The V1 and V2 segments are easily intervened upon granted the vessel is not tortuous or redundant [9]. The V2 segment typically has favorable anatomic features, including a short distance from the subclavian artery and a direct course through the transverse foramina of the cervical vertebra. Percutaneous intervention of the V3 segment is more difficult secondary to the mobility of the atlantoaxial and atlantooccipital joint, which results in extreme vessel tortuosity. Short self-expanding stents should be used in this region, and placement of balloon-expandable stents must be avoided for increased procedural success in the region due to its tortuosity. The V4 is the most difficult segment and is rarely attempted due to its high risk of major spinal cord complications and brain stem infarcts. The V4 segment intervention is reserved for acute stroke or severe symptoms unresponsive to medical therapy [5]. There are also several anatomic variants, including but not limited to: 1) anomalous left vertebral artery; 2) right carotid artery origin arising from the right carotid artery; 3) hypoplasia of a V.A. with congenital absence of V4 segment and termination in the posterior inferior cerebellar artery.

7.3. Arterial Access Considerations

Approximately 80% of VAIs are performed via the femoral artery. The ipsilateral brachial or radial artery is accessed in 20% of cases if needed for dual access during complex cases or in the setting of an acutely angulated proximal vertebral artery. A 5F or 6F pigtail catheter is used to perform an aortic arch angiogram in the left anterior oblique view to evaluate an anomalous vertebral artery. Typically, a 5F Berenstein diagnostic catheter is the catheter of choice. A 5F or 6F Judkins right, internal mammary, or Vitek curve catheter may also be used. In brachial or radial access, multipurpose catheters are the catheter of choice. The ideal interventional catheter should have a similar curve to the diagnostic imaging catheter. Typically, a 6F to 8F Judkins right or multipurpose guiding catheter can be used to perform most VAIs.

A 0.89-mm (0.035”) Guide wire or J-wire is used to advance the diagnostic catheter distal to the V.A. ostium. Engaging the V.A. ostium must be performed under continuous pressure monitoring. For VAI, a 0.014” soft guide wire is recommended with a Rapid Exchange (R.X.) or monorail balloon. Arterial sheaths can be removed once ACT < 180 sec.

7.4. Percutaneous Transluminal Angioplasty Balloons/Stents

The length of the balloon is dependent on lesion length and should cover the entire lesion. The balloon diameter should be 0.5mm less than that of the reference vessel. Adhering to the reference balloon rate, burst pressure is recommended to avoid complications, which is typically 6-8 atm. Consideration should be given to slight withdrawal of the balloon before post-stent high-pressure inflations up to 12 atm to prevent distal edge dissection. After balloon inflation, repeat angiography is performed to assess suboptimal lesion dilation, dissection, or perforation. All three findings are indications to proceed with V.A. stenting. Historically, aorto-ostial atherosclerotic lesions, in general, have a poor response to angioplasty alone due to elastic recoil. Given that the Vo and V1 segments are the most common regions for VAS and their close origin from the subclavian artery, they have a poor response to angioplasty alone, and hence, stenting should be considered [34]. In the V.A. ostium, balloon-expandable coronary or peripheral stents perform well on a 0.014 platform. To assure complete coverage of the ostium, stents are commonly extended by 1-2mm into the subclavian artery. In an isolated V1 or V2 segment that does not include the ostium, both self-expanding and balloon-expandable stents are acceptable.

Figs. (2A-2D) displays balloon angioplasty and stenting of a critically stenosed right VA.

Fig. (2A).

Ostial right vertebral artery stenosis (Arrow).

Fig. (2D).

Final result.

7.5. Bare Metal Stents (BMS) vs. Drug-Eluting Stents (DES)

Bare metal stents were used initially for VAI but reported restenosis rates ranging from 11%-43% [35, 36]. Ko et al. reported the use of DES for VAS with a high degree of technical success and relatively low incidence of peri-procedural complications; however, a high rate of in-stent restenosis was reported [37]. Similarly, in a single-center consecutive case series comparing DES with BMS for the treatment of VAS, Akins et al. demonstrated low complication rates and favored the use of DES [38]. Other studies comparing DES with BMS have also shown low restenosis rates of DES compared to BMS [39-41]. Additionally, a meta-analysis including 9 retrospective studies showed that DES restenosis rates were as low as 8.2% compared to 23.7% in BMS [42].

7.6. Intracranial vs. Extracranial Stenting

The benefits of intracranial and extracranial vertebral stenting differ according to anatomical considerations, including thinner walls of intracranial vessels that are at increased risk of perforation. Additionally, there is a risk of damaging the perforating arteries that arise from the intracranial vessels during stenting. In prior studies, the intervention of intracranial, when compared to extracranial, stenosis has been associated with a higher risk of early recurrent stroke and periprocedural stroke risk related to stenting [8, 43].

7.7. Post-Procedural Care

Dual antiplatelet therapy (DAPT) is recommended with Aspirin 81mg daily indefinitely and clopidogrel 75mg daily for 1 month in the setting of BMS placement. DAPT should be extended to at least 1 year with the use of DES. Post-procedure, patients are monitored in the hospital for 24 hours for possible neurovascular complications.

Duplex ultrasound of the vertebral arteries is performed post-procedure at 3, 6, and 12 months. If vertebrobasilar insufficiency (VBI) symptoms have resolved, the duplex ultrasound interval can be extended to annual examinations. Patients with recurrent VBI symptoms should undergo repeat angiography to identify restenosis and/or progression of atherosclerotic disease. All intracranial VAIs above the V3 segment are followed with a mandatory 1-year post-procedure selective angiography.

7.8. Innovation and Future Considerations

The interventional devices and medical regimens across the previously described VAI trials have varied widely over the past 30 years, some including only angioplasty rather than stenting. Advances in devices and techniques from peripheral artery to carotid artery interventions are being considered for the treatment of VAS. Such advances include the use of embolic protection devices (EPD), which can present a challenge in the treatment of VAS due to the tortuosity and smaller vessel caliber compared to carotid artery interventions.

However, a retrospective study by Jenkins et al. showed a very high procedural success rate of 100% without the use of distal EPD. Furthermore, the study showed durable symptom resolution in 80% of patients with few procedural complications (stroke rate 0% and TIA rate 1%) [5]. Interestingly, Canyigit et al. found an associated risk of asymptomatic distal embolization on diffusion-weighted MRI images in patients post VAI [44].

Geng et al. conducted a randomized trial comparing angiographic and clinical outcomes of self-expanding stents (SES) with distal EPD vs. balloon-expandable stents (BES) without EPD in the treatment of VAS and found a reduction in silent infarcts and in-stent restenosis in SES + EPD compared to BMS without EPD [45]. Presently, there are no U.S. Food and Drug Administration (FDA) approved embolic protection devices for VAIs. Although off-label, it is reasonable to use an embolic protection device for all VAIs with an adequate landing zone distal to the index lesion. If the operator elects to use an embolic protection device, the guide wire must always be kept within view, particularly when using a hydrophilic guide wire to decrease the risk of perforation and potentially fatal intracranial hemorrhage. Due to the risk of distal embolization of debris, debulking devices are not used for cerebral circulation.

DAPT, in addition to statin therapy, has been suggested to be more effective at preventing embolization in large-artery stroke and recurrent events after TIA and stroke [46, 47]. A randomized trial with randomization of medical therapy with DAPT compared to VAS stenting with embolic protection and DAPT has yet to be performed and would be a future consideration.

7.9. Procedural Complications

The major adverse events associated with VAI include TIA 1-2%, major stroke 1-2%, and death <1% [5, 8]. When endovascular intervention of VAS complicates with stroke or TIA, urgent repeat imaging is required to assess for potential mechanical complications related to VAI. Furthermore, in the setting of acute thromboembolic events, catheter-directed thrombolysis could be implemented. The rate of in-stent restenosis for VAI is variable based on prior studies and ranges between 0-43% [48-50]. Additional potential complications related to VAI are those that are inherent to invasive procedures and include but are not limited to death, access site bleeding, vessel rupture, and acute renal failure. A multidisciplinary team approach, including a neurologist, neuro-interventional radiologist, and cardiologist, in addition to the operator, is recommended to assist with complication management.

8. SURGICAL TREATMENT OF VERTEBRAL ARTERY OCCLUSION

The open surgical approach is more accessible when the lesions involve the V1 and V3 segments; however, the V2 and V4 segments are much more difficult to access.

Given the nature of this procedure with increased surgical risk, it should be performed in designated centers with experienced staff having appropriate expertise, as well as favorable outcomes.

8.1. Indications for Surgical Management

The decision to intervene should be based on:

1. Whether the patient is symptomatic or not.

2. Whether the symptoms are embolic, hemodynamic, or thrombotic.

3. The degree of stenosis, being 50-60 percent or more, in symptomatic patients.

4. In asymptomatic patients, stenosis of greater than 70 percent in the dominant V.A. or single V.A.

8.2. Surgical Methods

The location of the V.A. disease usually determines the type of intervention that is needed, with some exceptions. For instance, in stenosed ostial V1 lesions, transposition of the proximal V.A. to the carotid artery is the intervention of choice, and for more distal lesions, a bypass surgery from the common carotid artery to the V.A. between the first and the second cervical vertebrae is the commonest technique. Surgical reconstruction techniques that have been described for the treatment of proximal V.A. stenosis include endarterectomy of the V.A. as well as the subclavian artery, transposition of the V.A. to the subclavian or common carotid artery, venous graft linking bypass from the subclavian artery to the V.A., and the use of venous grafts from the common carotid artery or the subclavian artery to the V.A.

8.3. Complications

Reported post-operative complications include immediate thrombosis resulting in strokes with distal reconstruction having higher stroke and death (3-4%) rates than proximal V.A. surgeries, Vagus and recurrent laryngeal nerve palsy in proximal V.A. reconstruction (2%), Horner’s syndrome, Lymphocele, chylothorax.

9. SUMMARY OF RANDOMIZED CONTROLLED TRIALS FOR VERTEBRAL ARTERY INTERVENTION

The first randomized controlled trial to assess the benefit of endovascular treatment of VAS was the international, multicenter, Carotid and Vertebral Artery Transluminal Angioplasty Study (CAVATAS).

The CAVATAS trial failed to show the benefit of endovascular treatment of VAS, possibly due to the small number of patients (total 16) and outcome events, further limiting inferences on the relationship between stenosis severity and the risk of experiencing a recurrence of symptoms. Furthermore, the study was not able to compare the risk of restenosis after PTA alone with that after stenting because only 1 patient out of the 2 stented had follow-up imaging [31].

In the open-label, randomized Vertebral Artery Stenting Trial (VAST); during the initial 30-days in the intention-to-treat population, the primary composite outcome (vascular death, myocardial infarction, or any stroke) occurred in 3 of 57 patients in the stenting group (5%, 95% CI 0-11) and 1 of 58 patients in the OMT group (2%, 95% CI 0-5). All 4 of these events were vertebrobasilar strokes, one of them is fatal in the stenting group. Furthermore, in patients with intracranial stenosis, the primary outcome occurred in 2 of 9 patients in the stenting group and none of the 10 patients in the OMT alone. The 1-year event rates for stroke in the territory of the symptomatic VA were 9% in the stenting group (95% CI 2-16) and 7% (95% CI 0-13) in the OMT alone group. The VAST trial showed that about 1 in 20 patients had complications with a major peri-procedural vascular event, and the accumulative incidence of recurrent vertebrobasilar stroke in the OMT was relatively low [51-60].

The Vertebral Artery Ischemia Stenting Trial (VIST) was a multicenter, prospective, randomized, open-blinded endpoint clinical trial that evaluated patients with symptomatic VAS and randomly assigned them (1:1) to endovascular treatment plus OMT or OMT alone [61-66]. Randomization was further stratified by site of stenosis to extracranial (78.7%) versus intracranial (21.3%) and occurred in 179 patients, of which 91 were assigned to endovascular treatment plus OMT and 88 OMT alone.

The primary endpoint of stroke occurred in 5 patients in the endovascular intervention group (1 fatal stroke) and 12 patients in the medical group (2 fatal strokes), with an H.R. of 0.40 (95% CI 0.14-1.13, p= 0.08). The H.R. for patients with extracranial and intracranial VAS was 0.37 (95% CI 0.10-1.36) and 0.47 (05% CI 0.08-2.60), respectively. Although there was no significant difference in risk of stroke between the two groups, the low peri-procedural risk suggested that stenting is safe for extracranial stenosis [8]. Details of these randomized trials are summarized in Table 1.

Table 1.

Review of randomized control trial results of endovascular treatment for VAS.

RCT trial, Ref (Year)	n	Peri-procedure Complication	Any Stroke or Death within 30 Days	Stroke in Territory of Symptomatic VAS	Death	Any Stroke or TIA	Restenosis
CAVATAS32 (2007) Stent group Medical group	8	TIA (2)	0	2 (25%)	3/8	5/8	37.5%
	8	None	0	2 (25%)	3/8	2/8	N/A
VAST53 (2015) Stent group Medical group	57	Stroke (3)	3 (5%)	3 (5%)	1 (2%)	8/57	—
	58	None	1 (2%)	1 (2%)	2 (3%)	7/58	—
VIST8 (2017) Stent group Medical group	61	Stroke (3)	3 (5%)	3 (5%)	7 (11%)	9/61	—
	88	None	—	8 (9%)	9 (10%)	22/88	—
VAOS54 (2019) Stent group Medical group	43 55	None None	— —	— —	0/43 0/55	1/43 4/55	0% N/A

10. SUMMARY OF NON-RANDOMIZED CONTROL TRIALS FOR VERTEBRAL ARTERY INTERVENTION

Multiple case reports and case series have reported the successful use of endovascular procedures in the treatment of atherosclerotic disease of the posterior circulation [67-70]. Currently, the endovascular treatment of VAS remains a challenge due to a lack of large randomized controlled trials investigating the impact of optimal medical therapy and comparing it with VAS stenting or surgical repair. Irrespective of these challenges, a review of prior studies demonstrates that endovascular treatment of the V.A. is safe with high technical and clinical success rates and low clinical complication rates with stable long-term results. The durability of balloon angioplasty alone, BMS, and DES for VAS intervention is evidenced by low restenosis rates in multiple large case series reported despite restenosis rates ranging from 0% to 43% [71-74]. While there is a lack of data from randomized trials demonstrating the superiority of endovascular versus optimal medical therapy for the treatment of this disease, patients who have failed medical therapy should be considered for endovascular stenting of symptomatic VAS. Non-randomized studies on vertebral stenting are summarized in Table 2.

Table 2.

Review of results of endovascular stent for VAS.

Author (Year)	n	Technical Success Rate (%)	Procedural Complications	Improvement in Symptoms	Mean Follow- up(mo)	Late Stroke	Restenosis
Jenkins et al. (2001)	32	100	TIA (1)	31/32	10.6	0/32	1/32
Albuquerque et al. (2003)	33	97	CVA (1)	27/33	16.2	1/33	43%
Chastain et al. (1999)	50	98	None	48/50	25	1/50	10%
Lin et al. (2004)	58	100	CVA (3)	56/58	31.3	0/58	25%
Weber et al. (2005)	38	95	TIA (1)	23/26	11	0/26	36%
Cloud et al. (2003)	14	100	TIA (1)	13/14	33.6	1/14	36%
SSYLVIA Study Investigators, (2004)	18	100	None	—	6	2/18	43%
Jenkins et al. (2010)	112	100	TIA (1)	95/105	29	5/105	13%
Hatano et al. (2011)	117	99	TIA (2)	113/116	6	2/117	10%
Lin et al. (2006)	80	100	CVA (3)	78/80	12	0/80	28%
Karameshev et al. (2010)	10	100	TIA (1)	10/10	10	0/10	10%
Lin et al.,(2008)	11	100	None	11/11	8	0/11	0%
Zhou et al. (2011)	61	100	None	—	12	—	27%
Gupta et al. (2006)	31	100	None	31/31	4	0/31	7%
Vajda et al. (2009)	48	100	None	48/48	7	0/48	12%
Zaidat et al. (2015)	58	54	CVA (6)	—	12	7/50	29.4%

CONCLUSION

Treatment of patients with symptomatic VAS should focus on the reduction of vascular risk to prevent long-term stroke. Although there is insufficient evidence from randomized trials demonstrating the superiority of endovascular compared to optimal medical therapy for VAS, the durability and safety of the endovascular treatment is evidenced by low restenosis rates and procedural complications in multiple large series reported in the literature. Furthermore, VA stenting is a less morbid approach than open surgery for the treatment of symptomatic V.A. atherosclerotic obstructive disease; therefore, patients who fail medical therapy should be considered for endovascular stenting of symptomatic VAS.

Fig. (2B, 2C).

4.0 x 23mm stent positioned 2 mm proximal to vertebral artery ostium.

CONSENT FOR PUBLICATION

Not applicable.

FUNDING

None.

CONFLICT OF INTEREST

The authors declare no conflict of interest, financial or otherwise.