Article Text

Download PDFPDF

Current endovascular strategies for posterior circulation large vessel occlusion stroke: report of the Society of NeuroInterventional Surgery Standards and Guidelines Committee
  1. Yasha Kayan1,
  2. Philip M Meyers2,
  3. Charles J Prestigiacomo3,
  4. Peter Kan4,
  5. Justin F Fraser5
  6. on Behalf of the Society of NeuroInterventional Surgery
    1. 1 Neurointerventional Radiology, Abbott Northwestern Hospital, Minneapolis, Minnesota, USA
    2. 2 Neurointerventional Surgery, Columbia Presbyterian Hospital, New York City, New York, USA
    3. 3 Neurological Surgery, University of Cincinnati, Cincinnati, Ohio, USA
    4. 4 Department of Neurosurgery, Baylor College of Medicine, Houston, Texas, USA
    5. 5 Neurological Surgery, University of Kentucky, Lexington, Kentucky, USA
    1. Correspondence to Dr Justin F Fraser, Neurological Surgery, University of Kentucky, 800Rose St Room MS105A, Lexington, KY40536, USA; Jfr235{at}


    Background The aim of this publication is to provide a detailed update on the diagnosis, treatment, and endovascular techniques for posterior circulation emergent large vessel occlusion (pc-ELVO).

    Methods We performed a review of the literature to specifically evaluate this disease and its treatments.

    Results Data were analyzed, and recommendations were reported based on the strength of the published evidence and expert consensus.

    Conclusion While many questions about pc-ELVO remain to be studied, there is evidence to support particular practices in its evaluation and treatment.

    • thrombectomy
    • stroke
    • standards

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


    Posterior circulation strokes account for approximately 20% of all strokes, but posterior circulation emergent large vessel occlusions (pc-ELVO) are rare, representing only 1% of all ischemic strokes, and 5% of all ELVOs.1–3 These strokes are often devastating events. Good clinical outcomes occur in approximately 20% despite advanced care, and the rate of good clinical outcome after revascularization has been lower than for anterior circulation stroke.4–6 A number of reasons account for this disparity. Unlike hemispheric ischemia, basilar artery occlusion can mimic other clinical conditions, resulting in delays to clinical neurological evaluation and identification.7 There may be an extended prodrome that can potentially last days to months.8 9 Furthermore, the neurologic ‘real estate’ involved in brainstem ischemia is more eloquent; a small completed infarction can be clinically devastating. Etiologies for posterior circulation ischemia include thromboembolism (often in younger age), intrinsic atherosclerosis (usually sixth/seventh decades of life), dissection with basilar embolism (often in trauma), and vasculitidies.8 10 These factors all require consideration in the assessment of acute stroke referable to the posterior circulation.

    Basilar artery occlusion (BAO) syndrome presents as a subtype of posterior circulation stroke. While there is large practice variability among interventionalists on time limits for thrombectomy in this syndrome, good clinical outcomes have been demonstrated up to 50 hours from last known well.11 Stroke severity at presentation is an independent predictor of patient outcome.12 However, good outcomes appear to depend, at least in part, on revascularization of the occlusion.13 14 Lindsberg et al report good clinical outcomes with recanalization in 38% versus 2% without recanalization.15 From the Helsinki Stroke Thrombolysis Registry, Sairanen et al report good clinical outcomes with recanalization in 86% versus 14% without recanalization.7 Similarly, Kumar et al report 1.5 times decreased dependency and two times decreased mortality with recanalization than without it.14

    With these data in mind, the aims of this document include: (1) to review data on the natural history and outcomes from pc-ELVO and (2) to provide guidelines and recommendations on acute intervention regarding pc-ELVO (figure 1). Recommendations follow the American College of Cardiology/American Heart Association (ACC/AHA) classification of recommendation and level of evidence.16

    Figure 1

    Flow chart illustrating the workflow from the summary of recommendations. CTA, CT angiography; DWI, diffusion weighted imaging; IA, intra-arterial.


    Clinical presentation

    The clinical presentation of posterior circulation strokes can be non-specific. Common symptoms can include loss of consciousness, headache, nausea, vomiting, dizziness, double vision, hearing loss, slurred speech, vertigo, imbalance, and unilateral extremity weakness. Physical examination may also commonly reveal ataxia, nystagmus, and visual field defects.17BAO vary based on thrombus location. Top of the basilar syndrome includes somnolence, peduncular hallucinosis, convergence nystagmus, skew deviation, oscillatory eye movements, retraction and elevation of the eyelids, and vertical gaze paralysis. Mid-basilar occlusions can result in the various pontine syndromes. Proximal basilar occlusions can result in ‘locked in’ syndrome.17

    Prehospital ELVO scales

    Recent prehospital stroke scales were designed for anterior circulation strokes. In the evaluation of the Los Angeles Motor Scale, sensitivity and specificity in 119 patients to detect ELVO were 81% and 89%, respectively, but excluded posterior circulation strokes.18Rapid Arterial Occlusion Evaluation was found to have a similar ELVO sensitivity and specificity (85% and 65%, respectively) in a cohort of 357 patients, but only 7% had posterior circulation stroke.19 While it has recently been shown that adding the finger to nose test to the Cincinnati Prehospital Stroke Scale significantly increased the recognition of posterior circulation strokes by paramedics in a cohort of 789 consecutive ischemic stroke cases, pc-ELVO remains difficult to detect in the field.20

    Limitations of the NIHSS in pc-ELVO

    Limitations of the National Institutes of Health Stroke Scale (NIHSS) include a focus on limb and speech impairments and less emphasis on cranial nerve lesions.21 In a prospective cohort of 310 stroke patients not treated with thrombolysis, baseline NIHSS cut-off for a favorable 3 month outcome (modified Rankin Scale (mRS) score of ≤2) was significantly lower in patients with posterior circulation stroke compared with anterior circulation stroke.22 Indeed, patients with pc-ELVO can have an NIHSS score of 0, where symptoms may be headache, vertigo, and nausea only. The most common neurological sign in a series of 20 NIHSS 0, diffusion weighted imaging (DWI) positive patients was truncal ataxia.23 Therefore, the physician must maintain a high index of clinical suspicion for this condition in the face of subtle symptoms.

    Clinical confounders and differential diagnosis

    Posterior circulation strokes presenting to an emergency department may be associated with a delay in neurology evaluation, and door to needle time for intravenous tissue plasminogen activator (IV tPA) is often significantly longer compared with anterior circulation strokes. This is likely related to difficulty in timely recognition due to non-specific symptoms.7 The cause of BAO may not be immediately apparent on imaging. The differential diagnosis for etiology includes thromboembolism, cardiogenic or artery to artery embolism, or in situ thrombosis of vertebrobasilar stenosis due to atherosclerotic disease or, more rarely, dissection.

    In a Korean series of 62 patients with acute basilar occlusion who underwent stent retriever thrombectomy, 15 (24.1%) also underwent intracranial angioplasty and stenting for an underlying atherosclerotic stenosis. Proximal basilar occlusions were more likely to be related to atherosclerosis, and distal occlusions were more likely thromboembolic. Bilateral thalamic infarction on pretreatment DWI was less common in patients with atherosclerotic stenosis (0% vs 27.7%, P=0.027). Otherwise, patients with and without underlying atherosclerotic stenosis who underwent endovascular therapy had similar outcomes.24

    Natural history and outcomes with medical treatment

    Initially described in 1946, BAO was considered universally fatal.25 While there are limited data on BAO outcomes prior to the era of thrombolysis, data from 82 patients at three centers reported in 2005 show a case fatality of 40% and a poor outcome in almost 80% of patients.@@@ Zeumer et al first described successful basilar artery thrombolysis in 1982, and Hacke et al reported the clinical outcome benefit of revascularization in a series of 65 patients 6 years later in 1988.26 27 Advancement in non-invasive imaging permitted understanding of variances in outcome. The New England Stroke Registry showed that 29% of patients with posterior circulation strokes died or suffered severe disability; however, individuals with embolic strokes fared worse than those with in situ atherosclerosis.8 However, reported mortality ranges from 45% to 86%.4 15 28 Predictors of poor outcome include older age, higher NIHSS score, lack of recanalization, atrial fibrillation, intracranial hemorrhage, and posterior circulation Alberta Stroke Program Early CT Score (pc-ASPECTS) ≤8.29

    As with anterior circulation strokes, hemorrhagic conversion is related to the overall volume of the infarction, use of anticoagulants (eg, heparin) in the peri-ictal setting, and administration of fibrinolytics. In a small study that reviewed 37 patients with posterior circulation strokes who were administered heparin, 11% developed hemorrhagic conversions. All patients who worsened sustained a proximal to mid-basilar artery occlusion and cerebellar infarction.30 In a retrospective review of almost 1000 patients in the TIMS-China registry, multivariate analysis suggested a lower 3.2% incidence of symptomatic hemorrhagic conversion in 124 patients with posterior circulation strokes compared with 7.7% in patients with anterior circulation strokes.31 Moreover, a higher proportion of functional independence was noted after hemorrhagic conversion in patients with posterior circulation strokes when compared with patients with anterior circulation strokes (64% vs 53%, OR 2.33, 95% CI 1.40 to 3.89).


    Computed tomography

    Non-contrast CT is used to exclude hemorrhage and evaluate for other intracranial abnormalities, including life-threatening edema and mass effect in the posterior fossa. Beam hardening artifact often limits assessment of structures in the posterior fossa but vessel hyperdensity, particularly a hyperdense basilar artery, is sometimes present.32 In a series of 95 patients with posterior circulation infarction who underwent CT brain scan with CT angiography (CTA), 14 had partial or complete BAO. Of the 14, 10 (83%) had a hyderdense basilar artery on non-contrast CT. A hyperdense basilar artery was a significant independent predictor of a poor outcome at 6 months (mRS >2, OR 5.6, 95% CI 1.1 to 33.3). A hyperdense basilar artery had a sensitivity of 71% and specificity of 98% as a predictor of basilar occlusion.33 Diagnostic accuracy may be improved by measuring vessel density (optimal cut-off is 40–42 Hounsfield units).34

    CTA has become the primary method used to identify large vessel occlusion in most stroke patients.35–39 Because posterior circulation strokes were excluded from the major randomized controlled trials for thrombectomy, the evidence for CTA in the posterior circulation is limited but the logic remains compelling. In addition to identifying the site of occlusion, it can have prognostic significance. In a retrospective review of 15 patients with vertebrobasilar stroke treated with thrombectomy, a 6 point posterior circulation CTA vascular collateral score as well as patency of the distal third of the basilar artery correlated with good outcomes, defined by mRS score of ≤3 at 3 months.40 Several studies, including a series of 21 patients and another of 104 patients, found that the presence of bilateral posterior communicating arteries on pretreatment CTA was associated with more favorable outcomes after endovascular recanalization.41 42 In the Basilar Artery International Cooperation Study (BASICS) study, evaluation of comatose patients with basilar occlusions using CTA showed the extent of brainstem ischemia on source images was related to mortality and poor outcome.43 The use of CTA in patients with possible BAO has also been found to be cost effective.44

    Historically, the use of CT perfusion (CTP) in the posterior circulation has been of limited value. However, the prospective, multicenter, observational Dutch Acute Stroke Study (DUST) of 88 patients with suspected posterior circulation stroke found that CTP adds additional diagnostic value, primarily in the form of higher sensitivity (74%) and negative predictive value (80%) compared with non-contrast CT (31% and 61%, respectively) and CTA (33% and 62%, respectively). Large vessel occlusions were not specifically evaluated.45 Another study found no difference between detection of infratentorial and supratentorial lesions using CTP; however the duration of stroke onset to imaging was long (540 min), likely favoring identification of ischemic changes, and thus sensitivity, with non-enhanced CT alone.46 In BASICS, a small minority of patients were evaluated using CTA and CTP (27 of 592, 4.6%). Mean transit time was abnormal in 93%. Cerebral blood volume correlated with a risk of death. For three patients with pc-ASPECTS <8, all had died at 1 month compared with 6 of 23 patients with a cerebral blood volume pc-ASPECTS ≥8 (RR 3.8, 95% CI 1.9 to 7.6).47

    Magnetic resonance imaging

    In ENDOSTROKE, an observational registry that included 148 consecutive patients with basilar occlusion who underwent mechanical thrombectomy, the use of MRI prior to intervention was an independent predictor of good clinical outcome.6 Spin-echo MRI can be used to evaluate for potential edema and mass effect in the posterior fossa, although non-contrast CT is more readily obtained at most medical centers in the acute setting. The extent of signal hyperintensity of the basilar artery on FLAIR imaging has been associated with mortality in a retrospective review of 20 basilar occlusion patients.48 DWI may be the most sensitive test for evaluation of ischemia but can initially be normal in 6–10% of cases, twice as often as DWI in the anterior circulation.49–51

    pc-ASPECTS, originally described using CTA source images,52 may also be calculated using DWI. For reference, pc-ASPECTS assigns 10 points to the entire posterior circulation, with 1 point each subtracted for early ischemic change in the right or left thalamus, cerebellum, or posterior cerebral artery territories (total 6 points), and 2 points each for early ischemic change in any part of the midbrain or pons (total 4 points). The extent of restricted diffusion used to calculate pc-ASPECTS was an independent predictor of unfavorable outcome (mRS ≥3, OR 0.40, 95% CI 0.23 to 0.67) in a retrospective study of 132 patients.52 53 Nagel et al reported a retrospective study of 50 patients with acute basilar occlusion, showing that the DWI pc-ASPECTS ≥8 was the only independent predictor for favorable outcome (OR 3.9, 95% CI 1.4 to 11.7).54 In 35 patients who underwent mechanical thrombectomy for basilar occlusion using MRI based selection criteria, DWI pc-ASPECTS was significantly different between those patients with a good outcome (mRS 0–2) compared with patients who suffered a poor outcome (7.8±1.6 vs 5.4±1.8, P=0.001).55 DWI pc-ASPECTS has also been found to be useful in predicting outcome in patients with acutely symptomatic basilar stenosis and not occlusion.56

    There is little in the literature specifically looking at MR perfusion in the posterior circulation. In five patients with acute basilar occlusion receiving intra-arterial thrombolytics, significant diffusion–perfusion mismatch (mean 73%, range 49–99%) was visualized in all patients with pretreatment imaging, suggesting the presence of salvageable tissue. Final infarction volumes were smaller than pretreatment perfusion volumes, suggesting tissue salvaged by thrombolysis.57

    CT and CTA are recommended in the evaluation of patients with suspected pc-ELVO, with possible DWI for late arrivals, as institutionally defined, to exclude those with a poor pc-ASPECTS or large core infarction (AHA Class I, Level of Evidence C-LD).

    Acute patient care

    Guidelines applicable to ischemic strokes of the anterior circulation apply when managing patients with pc-ELVO.58 As in any acute care setting, careful and frequent assessment of the patient’s airway, breathing, and circulation is required, particularly given the potential fluctuating nature of posterior circulation strokes. Due to the brain territory involved, patients with pc-ELVO are at high risk for loss of consciousness, airway compromise, respiratory depression, and autonomic instability.


    In a series of 153 patients with stroke complicated by hypoxemia, Rowat et al demonstrated increased mortality and a greater risk of post-stroke hypoxemia.59 Loss of sensation to the pharynx and hypopharynx, along with reduction in reflexes, increases the risk of airway compromise.60 Central breathing abnormalities and associated hypoxia have also been shown to be frequent complications of posterior circulation stroke.61

    Given poorer outcomes in stroke patients with hypoxia, administration of supplemental oxygen to maintain oxygen saturation above 94% may be useful.58 While the least invasive route for oxygen supplementation is preferred, in the setting of potential airway compromise or focal brainstem related deficits, endotracheal intubation and ventilation is warranted. Given the aspiration risk, this may reduce the number of hospital acquired pneumonias and potentially aid in the management of intracranial pressure.62 63 Unfortunately, it is important to note that the outcome for patients who undergo intubation is poor.64

    Although there is some evidence that patients with anterior circulation strokes and no evidence of hypoxia can benefit from being placed in the supine or decubitus position, the high risk of aspiration in patients with pc-ELVO would preclude placing the patient in the supine position.65 Thus in patients with a risk of elevated intracranila pressure (ICP) or aspiration risk, the head of the bed should be elevated to between 15 and 30°.66


    Posterior circulation ELVO patients should have frequent, if not continuous, blood pressure monitoring as it serves as a potential indicator of imminent ICP elevation. There is a delicate balance between blood pressure adequate to perfuse the brainstem and increased cardiac work resulting in myocardial ischemia or elevated ICP. Studies have demonstrated improved outcomes in patients whose systolic blood pressure ranged from 121 to 200 mm Hg and diastolic blood pressures ranging from 81 to 110 mm Hg.67–69 The current recommendation not to lower blood pressure within the first 24 hours of a stroke unless it exceeds 220/120 mm Hg or there are concomitant medical indications to do so, is reasonable.58


    Hyperglycemia is common in the acute ischemic stroke population, with some studies reporting an elevated admission blood glucose level in >40%.70 There are no data with regards to hyperglycemia specifically in posterior circulation stroke. Poor clinical outcomes have been reported in patients with elevated blood glucose levels in the setting of symptomatic intracerebral hemorrhage after intravenous fibrinolysis using recombinant tPA, and increase in infarction volume as documented on MRI.70 A randomized trial to evaluate the effects of hyperglycemia in stroke failed to reach a conclusion as the trial was stopped early.71 At present, it is recommended that blood sugar be maintained between 140 and 180 mg/dL in accordance with the American Diabetic Association guidelines for hospitalized patients. It is also suggested that subcutaneous insulin protocols can be used safely and effectively in patients with stroke without the need for intravenous insulin protocols.58


    No specific data exist for the incidence of hyperthermia (>37.6°C) in patients presenting with posterior circulation stroke.72 However, hyperthermia in the setting of acute stroke has been associated with poor neurological outcome.73 A recent meta-analysis demonstrated a twofold increase in short term mortality in hyperthermic patients within 24 hours of ictus.74 Administration of aspirin or acetaminophen can achieve normothermia in patients presenting with temperatures <38°C.75 However, hypothermia protocols in the setting of pc-ELVO as a neuroprotectant have not been assessed.

    General guidelines for the acute medical management of patients with ischemic stroke also apply to the subgroup of pc-ELVO patients (AHA Class I, Level of Evidence C-LD).

    Surgical considerations

    The most significant surgical concern in patients presenting with pc-ELVO is the immediate and delayed risk of herniation secondary to cerebellar infarction or hemorrhage. The small volume of the posterior fossa and the eloquence of brainstem structures contribute to the risk of catastrophic herniation and fulminant clinical deterioration, respectively. Mass effect on the brainstem by an infarcted cerebellar hemisphere or non-communicating hydrocephalus secondary to occlusion of the cerebral aqueduct may also result in a fluctuating neurological examination and require repeat imaging.

    Although mass effect can peak on the third day post-infarct, it has been reported to occur throughout the first week. Approximately 25% of patients will develop mass effect causing rapid clinical deterioration.76 77 Ventricular or aqueductal obstruction causes hydrocephalus in up to 20% of patients with cerebellar stroke.76 Thus it is imperative to identify and aggressively treat patients with evidence of mass effect and hydrocephalus, as 85% of patients progressing to coma die without intervention.78

    Conservative measures such as elevating the head of bed, the use of osmotic diuretics, and hyperventilation provide only transient benefit. There is no consensus on the treatment of patients with hydrocephalus in this setting. Placement of an external ventricular drain alone carries a risk of upward herniation as well as continued mass effect on the brainstem.78 Thus suboccipital craniectomy, durotomy, and duraplasty to reduce the mass effect on the brainstem and restore CSF drainage should be considered early for these patients. Half of patients progressing to coma treated with suboccipital decompression have good outcomes. It has been demonstrated in several independent case series that rapid surgical intervention in the setting of acute clinical deterioration can improve outcome.79Randomized controlled trials addressing surgery in this patient population are unlikely to be performed due to the potential for herniation in a control group. In a rare prospective study on this topic, preoperative Glasgow Coma Scale was a major predictor of good clinical outcome.80 As such, early neurosurgical consultation should be sought for surgical decompression before clinical deterioration occurs.

    Emergent surgical decompression in patients with cerebellar infarct exhibiting mass effect with life threatening sequela is reasonable(AHA Class IIa, Level of Evidence C-LD).

    Endovascular management

    Mechanical revascularization

    There are no randomized controlled trials of mechanical thrombectomy for acute stroke due to BAO. A retrospective analysis of 34 consecutive patients undergoing thrombectomy for acute BAO showed that the Solitaire stent retriever and the ACE reperfusion catheters had a higher recanalization rate (92%) compared with older devices (24%). The procedures were also faster with the newer devices (88±31 min vs 126±58 min).81 Successful recanalization has been found not to depend on thrombus length and can be achieved despite a high clot burden.82 While mechanical thrombectomy in the basilar artery has been associated with more procedural complications than in anterior circulation strokes,83 failure of successful recanalization was a strong predictor of mortality in a prospective registry of 117 consecutive patients with BAO treated with either stent retrievers or large bore distal aspiration catheters. In the setting of recanalization, a lower pc-ASPECTS (OR 1.71, 95% CI 1.19 to 2.44) and angioplasty/stenting of the basilar artery (OR 4.71, 95% CI 1.34 to 16.54) were independent predictors of mortality after thrombectomy.84 A meta-analysis of 31 studies summarizing 1358 patients with basilar occlusions treated with fibrinolysis (either intravenous or intra-arterial recombinant tPA) and/or mechanical thrombectomy found data insufficient to generate evidence based medical recommendations; however, the mechanical thrombectomy group had a significantly higher rate of independence (mRS 0–2) than fibrinolysis without mechanical thrombectomy at 90 days (P<0.001).85

    Emergent mechanical thrombectomy is reasonable for pc-ELVO patients in order to maximize the chance of a good clinical outcome (AHA Class IIa, Level of Evidence B-NR).

    Suction thrombectomy

    A retrospective review of 436 acute ischemic stroke patients who underwent mechanical thrombectomy using A Direct Aspiration First Pass Technique (ADAPT86) included 57 (13%) patients with vertebrobasilar occlusions. The posterior circulation group had a similar likelihood of good outcome compared with the anterior group (42.9% vs 43.2%, respectively).87 In a prospective study of 100 patients with BAO treated with mechanical thrombectomy, successful reperfusion was a strong predictor of favorable outcome at 90 days (OR 4.57, 95% CI 1.24 to 16.87, P=0.023). Of the 100 patients, 46 patients were treated using ADAPT and 54 patients primarily with stent retrievers. ADAPT achieved a significantly higher rate of complete reperfusion (OR 2.59, 95% CI 1.14 to 5.86, P=0.021) with shorter mean procedure duration (median 45 min, IQR 34 to 62 min vs 56 min, IQR 40 to 90 min; P=0.05). The rate of periprocedural complications was also lower with ADAPT (4.3% vs 25.9%, P=0.003).84

    Suction thrombectomy is a reasonable form of mechanical thrombectomy for the emergent treatment of pc-ELVO patients (AHA Class IIa, Level of Evidence B-NR).

    Stent retriever thrombectomy

    Several retrospective series have examined the use of stent retrievers to treat BAO. A retrospective review of 161 consecutive stent retriever thrombectomy patients comparing 24 basilar occlusions with 137 anterior circulation occlusions found a significantly higher 90 day mRS score in the basilar group (4.2 vs 3.0, P=0.003). Mortality was also higher in the basilar group (16.6% vs 5.8%, P=0.001). Nonetheless, among the patients with BAO, successful recanalization was associated with a more favorable mRS at 90 days (P=0.027).88 In 50 consecutive patients who underwent stent retriever thrombectomy for acute basilar occlusion, multivariate analysis of factors potentially related to prognosis found that only low initial NIHSS score (OR 0.82, 95% CI 0.709 to 0.949, P=0.008) and high DWI pc-ASPECTS (OR 1.854, 95% CI 1.012 to 3.397, P=0.045) were significant independent predictors of good outcome (mRS 0–2) at 90 days12. In 69 patients treated with Solitaire AB thrombectomy for acute posterior circulation stroke caused by large intracranial vessel occlusion, stroke subtype (intracranial atherosclerotic disease vs embolism, OR 0.101, 95% CI 0.020 to 0.501, P=0.005) and pc-ASPECTS on DWI (≥6 vs <6, OR 7.335, 95% CI 1.495 to 36.191, P=0.014) were independent predictors of good clinical outcome at 90 days.89

    Stent retriever thrombectomy is a reasonable form of mechanical thrombectomy for the emergent treatment of pc-ELVO patients (AHA Class IIa, Level of Evidence B-NR).

    Angioplasty and stent angioplasty

    In situ thrombosis of an underlying atherosclerotic stenosis is the cause of BAO in some patients, most commonly in Asian populations. For symptomatic intracranial stenosis due to atherosclerotic disease, both Stenting versus Aggressive Medical Therapy for Intracranial Arterial Stenosis (SAMMPRIS) and Stenting of Symptomatic Atherosclerotic Lesions in the Vertebral or Intracranial Arteries (SSYLVIA) trials failed to show superior results with stent angioplasty compared with medical therapy alone.90 91 However, these studies looked at secondary stroke prevention in typically less acute settings. For acute BAO presenting with severe neurological deficits, the risk of major morbidity and death are high, and angioplasty and stenting of an associated plaque may be lifesaving. This is particularly true when there is flow limitation or thrombectomy related endothelial injury resulting in poor reperfusion or a high risk of re-occlusion.92

    Coronary and intracranial balloon catheters and stents have been applied anecdotally to the treatment of refractory acute cerebral artery occlusion with stroke with some success since 1987.16 In a series of 13 patients at a single Chinese center who underwent combined mechanical thrombectomy with angioplasty and stenting, 46% had a good functional outcome, with an mRS score of at least 2 at 90 days.92 In the anterior circulation, for patients with acute atherosclerotic occlusion, it has even been suggested that primary angioplasty and stenting may be superior to thrombectomy.93 Both intracranial angioplasty/stenting and intra-arterial infusion of a glycoprotein IIb/IIIa inhibitor were found to be safe and effective in treating underlying severe atherosclerotic stenosis in acute stroke patients with ELVO in a series of 140 patients in Korea, although only 1 in 5 patients in this study had basilar occlusions.94 Also, a higher stroke complication rate has been noted in angioplasty and stenting of perforator bearing arteries in the posterior circulation versus the anterior circulation. 95

    Angioplasty and stenting may be considered for pc-ELVO if there is a persistent severe stenosis following thrombectomy, particularly if there is poor reperfusion or a perceived high risk of re-occlusion(AHA Class IIb, Level of Evidence C-EO).

    Intra-arterial fibrinolysis

    Case series and the available trial data point to the benefits of revascularization in appropriately selected patients, but selection of patient likely to benefit remains a challenge. Prior to 2006, all reports of basilar recanalization were case series. The Australian Urokinase Trial enrolled 16 patients with BAO but was halted prematurely, and showed no overall treatment benefit, although recanalization was associated with good outcome (38% vs 2%).96 Meta-analysis of uncontrolled studies showed no benefit to treatment.15 In Interventional Management of Stroke 3 (IMS-3), five patients enrolled had vertebrobasilar occlusions, but there were no outcome differences in this small group.97 In the ENDOSTROKE multicenter registry, 148 consecutive patients with BAO were enrolled. By this time, stent retrievers were available for stroke treatment. Thirty per cent received some form of intra-arterial fibrinolysis in combination with mechanical methods. In addition, 84% of patients underwent thrombectomy using first generation thrombectomy devices, including the concentric MERCI retriever or Penumbra suction catheter. Overall, 34% achieved good outcome. The mortality rate was 35%. Lower NIHSS predicted good outcome, but revascularization did not. Thrombolysis in Cerebral Infarction 2b–3 recanalization was achieved in 79% but did not predict good outcome.6 Similarly, the BASICS study did not identify the superiority of intra-arterial treatment over intravenous fibrinolysis in the 619 patients entered into the registry.5

    Subsequently, between 2006 and 2015, the BASICS trial enrolled 38 consecutive patients with BAO to intra-arterial treatment; 71% also received intravenous fibrinolysis before intra-arterial therapy. Mechanical thrombectomy was performed in 30 patients (84%) while 7 patients (18%) received intra-arterial urokinase without thrombectomy. Adequate recanalization was achieved in 34 patients (89%) while good clinical outcome occurred in only 19 patients (50%). There was no association between good outcome and patent collateral circulation in the BASICS trial. Symptomatic intracranial hemorrhage occurred in two patient (5%). The authors concluded that the rate of good outcomes is comparable with the intra-arterial treatment group in the Multicenter Randomized Clinical trial of Endovascular treatment for Acute ischemic stroke in the Netherlands (MR CLEAN) trial. 98

    Intra-arterial pharmacologic thrombolysis may be considered for pc-ELVO if mechanical revascularization fails(AHA Class IIb, Level of Evidence C-LD).

    Most commonly, cerebrovascular procedures, including mechanical thrombectomy, are performed using a transfemoral approach. Recently, a series of nine cases of basilar thrombectomy from a radial approach were reported with technical success, defined as Thrombolysis in Cerebral Infarction  2b or 3, achieved in 89% of cases. Average puncture to revascularization time was 35.8 min. There were no complications related to radial artery catheterization. The authors advocate for radial access as a firstline approach for basilar thrombectomy given the anatomical advantages.99 There is little in the literature specifically for brachial access in the setting of pc-ELVO.100 Disadvantages of transbrachial or transradial access include the theoretical possibility of having to use smaller catheters as well as less operator familiarity with the technique in an acute setting.

    For the endovascular management of pc-ELVO, transbrachial or transradial access may be considered as an alternative to the traditional transfemoral route given anatomic advantages(AHA Class IIb, Level of Evidence C-EO).

    Summary of recommendations

    • CT and CTA are recommended in the evaluation of patients with suspected pc-ELVO, with possible DWI for late arrivals, as institutionally defined, to exclude those with a poor pc-ASPECTS or large core infarction (AHA Class I, Level of Evidence C-LD).

    • General guidelines for the acute medical management of patients with ischemic stroke also apply to the subgroup of pc-ELVO patients (AHA Class I, Level of Evidence C-LD).

    • Early surgical decompression in patients with cerebellar infarct exhibiting mass effect with life threatening sequela is reasonable (AHA Class IIa, Level of Evidence C-LD).

    • Emergent mechanical thrombectomy is reasonable for pc-ELVO patients in order to maximize the chance of a good clinical outcome (AHA Class IIa, Level of Evidence B-NR).

    • Suction thrombectomy is a reasonable form of mechanical thrombectomy for the emergent treatment of pc-ELVO patients (AHA Class IIa, Level of Evidence B-NR).

    • Stent retriever thrombectomy is a reasonable form of mechanical thrombectomy for the emergent treatment of pc-ELVO patients (AHA Class IIa, Level of Evidence B-NR).

    • Angioplasty and stenting may be considered for pc-ELVO if there is a persistent severe stenosis following thrombectomy, particularly if there is poor reperfusion or a perceived high risk of re-occlusion (AHA Class IIb, Level of Evidence C-EO).

    • Intra-arterial pharmacologic thrombolysis may be considered for pc-ELVO if mechanical revascularization fails (AHA Class IIb, Level of Evidence C-LD).

    • For the endovascular management of pc-ELVO, transbrachial or transradial access may be considered as an alternative to the traditional transfemoral route given anatomic advantages (AHA Class IIb, Level of Evidence C-EO).



    • Contributors YK, PMM, and CJP were the primary authors and, as such, were responsible for overseeing construction of the document outline, gathering of evidence, and drafting of the manuscript. PK acted as a representative of the Board of the Society of NeuroInterventional Surgery, and provided commentary and editing on behalf of the Board and of the Society as a whole. JFF, as senior author, was responsible for organizing the writing group, overseeing the outline construction, facilitating communication between the writing group and the Standards and Guidelines Committee, drafting and editing of the manuscript, and the submission/proofing for publication.

    • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial, or not-for-profit sectors.

    • Disclaimer This literature review (’Review') is provided for informational and educational purposes only. Adherence to any recommendations included in this review will not ensure successful treatment in every situation. Furthermore, the recommendations contained in this review should not be interpreted as setting a standard of care, or be deemed inclusive of all proper methods of care nor exclusive of other methods of care reasonably directed to obtaining the same results. The ultimate judgment regarding the propriety of any specific therapy must be made by the physician and the patient in light of all the circumstances presented by the individual patient, and the known variability and biological behavior of the medical condition. This review and its conclusions and recommendations reflect the best available information at the time the review was prepared. The results of future studies may require revisions to the recommendations in this review to reflect new data. SNIS does not warrant the accuracy or completeness of the review and assumes no responsibility for any injury or damage to persons or property arising out of or related to any use of this review or for any errors or omissions.

    • Competing interests JFF is an equity interest holder for Fawkes Biotechnology, LLC, and a consultant for Stream Biomedical and Medtronic Neurovascular. PK is a consultant for Stryker and Medtronic Neurovascular. YK is a consultant for Penumbra and for Medtronic Neurovascular. PMM is a consultant for Stryker, Medtronic, Penumbra, and Siemens. A S Arthur is a consultant for Balt, Johnson and Johnson, Leica, Medtronic, Microvention, Penumbra, Scientia, Siemens, and Stryker. He has research support from Cerenovus, Microvention, Penumbra, and Siemens. He is a shareholder in Bendit, Cerebrotech, Endostream, Magneto, Marblehead, Neurogami, Serenity, Synchron, Triad Medical, and Vascular Simulations. B Baxter is a consultant for Penumbra, Medtronic, Stryker, Metactive, and 880 Medical. He has stock in Penumbra, and stock options in He has ownership interest in Route 92 and Marblehead. Dr Chen is a consultant for Medtronic, Stryker, Genentech, Imperative Health, Microvention, and Cerenovus. G Dabus is a consultant for Medtronic, Microvention, Penumbra, Stryker, and Cerenovus. He is a shareholder in InNeuroCo. K Fargen is a consultant for Cerebrotech. D Frei is a consultant for Genentech, Penumbra, and Stryker. He has stock ownership in Penumbra. S Hetts has a Royalty Agreement with Penumbra. He contracts for Core Imaging Lab for Stryker and MicroVention. He has a research contract with Siemens. He is on Data Safety and Monitoring Boards for DAWN, ARISE2, and PHIL trials. He has grant support through NCI and NIBIB. M S Hussain is a consultant for Cerenovus. M V Jayaraman is a speaker for Medtronic. R P Klucznik is a consultant for Medtronic and Microvention. W J Mack is a consultant for Rebound Therapeutics, Viseon TSP, Medtronic, Penumbra, and Stream Biomedical. He is an investor in Cerebrotech, Endostream, Viseon, and Rebound Therapeutics. R McTaggart is a consultant for Stryker. J Milburn is a consultant for Stryker and Penumbra. M Mokin is a consultant for Canon Medical, Cerebrotech, and Imperative Care. J Mocco receives research support from Stryker, Penumbra, Medtronic, and Microvention. He is a consultant for Imperative Care, Cerebrotech, Viseon, Endostream, Rebound Therapeutics, and Vastrax. He is an investor/shareholder in BlinkTBI, Serenity, NTI, Neurvana, and Cardinal Consulting. G L Pride is a consultant for Sequent and Microvention-Terumo. He is on the Data Safety and Monitoring Board for Web-IT Study (Cerenovus) and for the Pulserider NAPA trial. R M Starke receives research support from NREF, Joe Niekro Foundation, Brain Aneurysm Foundation, Bee Foundation, NIH, Miami Clinical and Translational Science Institute, National Center for Advancing Translational Sciences, and the National Institute on Minority Health and Health Disparities. He is a consultant for Penumbra, Abbott, Medtronic, and Cerenovus. P J Sunenshine is a consultant for Medtronic. G Toth is a consultant for Dynamed EBSCO.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Collaborators A S Arthur, S Ansari, B Baxter, K Bulsara, G Dabus, K Fargen, D Frei, C D Gandhi, S Hetts, M S Hussain, M V Jayaraman, R P Klucznik, S K Lee, W J Mack, R McTaggart, T Leslie-Mazwi, J Milburn, M Mokin, J Mocco, A T Patsalides, G L Pride, R M Starke, P J Sunenshine, G Toth, Michael Chen.

    • Patient consent for publication Not required.