Recommendations

1. The availability of intra-arterial therapy should generally not preclude the intravenous administration of recombinant tissue plasminogen activator (rt-PA) in otherwise eligible patients (American Heart Association (AHA) Class I, Level of Evidence A, Centre for Evidence Based Medicine (CEBM) level 1b, grade A–B).*

2. Treatment requires the patient to be at an experienced stroke center with rapid access to cerebral angiography and qualified interventionalists (AHA Class I, Level of Evidence C, CEBM level 2C, grade C).*

3. Intra-arterial chemical thrombolysis is an option for the treatment of selected patients with major stroke of <6 h duration due to an occlusion of the middle cerebral artery and who are not otherwise candidates for intravenous rt-PA (AHA Class I, Level of Evidence B, CEBM level 2b, grade B).*

4. Intra-arterial chemical thrombolysis is reasonable for patients who have contraindications to the use of intravenous thrombolysis, such as recent surgery (AHA Class IIa, Level of Evidence C, CEBM level 4, grade C).*

5. Combination intravenous/intra-arterial therapy is reasonable in selected patients who present with major stroke of <4.5 h duration (AHA Class IIa, Level of Evidence B, CEBM level 2a, grade B).

6. Intra-arterial thrombus removal with the Penumbra aspiration system or Concentric MERCI clot retrieval device is reasonable in selected patients with major stroke where care has been initiated at <8 h duration although data regarding improvement of clinical outcomes is unclear (AHA Class IIa, Level of Evidence B, CEBM level 2a, grade B).

7. The usefulness of other endovascular devices is not yet established, but they may be beneficial and may be considered (AHA Class IIb, Level of Evidence C, CEBM level 4, grade C.)*

8. The usefulness of endovascular treatment in the posterior circulation is not yet established, but it may be beneficial and may be considered, even beyond the 6–8 h time window typical for anterior circulation stroke (AHA Class IIB, Level of Evidence C).

*As already defined by the AHA guidelines for the early management of adults with ischemic stroke (Circulation 2007) and AHA statement regarding indications for the performance of intracranial endovascular neurointerventional procedures (Circulation 2009).1 2

Intravenous thrombolysis

The availability of intra-arterial therapy should not preclude the intravenous administration of recombinant tissue plasminogen activator (rt-PA) in otherwise eligible patients (AHA Class I, Level of Evidence A, CEBM level 1b, grade A–B).

At present, the only therapy demonstrated to improve clinical outcomes from AIS in randomized controlled clinical trials is thrombolysis of the clot responsible for the ischemic event.22–25 Specifically, since 1996, intravenous (IV) thrombolysis with recombinant tissue plasminogen activator (rt-PA) is FDA approved for the treatment of ischemic stroke to patients without evidence of hemorrhage seen on initial non-contrast head CT, if treatment is initiated within 3 h of a clearly defined symptom onset. A 2009 scientific advisory from the AHA/American Stroke Association26 recommended extension of the time window for treatment with IV rt-PA to 4.5 h, largely based on the prospective, randomized, placebo controlled European Co-operative Acute Stroke Study III (ECASS III) trial.23 rt-PA is administered intravenously at a dose of 0.9 mg/kg (maximum of 90 mg) with 10% of the total dose given as a bolus with the remainder infused over 60 min.26 27 This dosage has been shown to be safe and effective in routine clinical use, with results comparable with pooled, randomized, controlled trial data in the Safe Implementation of Thrombolysis in Stroke—Monitoring Study (SITS-MOST), a large observational registry.28 As noted below in the section on combination therapy (IV plus IA rt-PA), several trials and case series have shown that patients with a large clot burden are not likely to achieve recanalization with IV rt-PA. A 2011 retrospective review of 138 patient treated with IV rt-PA demonstrated that of the 76 patients who did not achieve recanalization, the majority had clot lengths which exceeded 8 mm. Clot length was determined via automated measurements of vascular hyperdensity on thin section pretreatment CT scans. The remaining patients who did achieve recanalization had clot lengths of <8 mm.29

Endovascular treatment

Treatment requires the patient to be at an experienced stroke center with rapid access to cerebral angiography and qualified interventionalists (AHA Class I, Level of Evidence C, CEBM level 2c, grade C).

Endovascular treatment of stroke requires the infrastructure to support the rapid assessment, stabilization and transport of patients with AIS, along with documented procedures for communication with the receiving hospital emergency room and stroke team. Evaluation of the patient before and after the procedure requires a multidisciplinary team approach not only for discussion of endovascular treatment options, particularly on full evaluation of the underlying vascular disease, but also for evaluation and treatment of the attendant complications. Nursing and technological staff should have training in endovascular procedures and should be available on a continuous basis with a 60 min response time to the hospital. There should be documentation and evaluation of procedural indications, outcomes and complications with the support of a quality assurance program. This brief description of an experienced stroke center is subsumed within the more detailed evidence based guidelines describing the concept of a Comprehensive Stroke Center (CSC), defined by the Brain Attack Coalition in 2005 as a facility or system that is able to “diagnose and treat stroke patients who require a high intensity of medical and surgical care, specialized tests or interventional therapies”.30 The CSC is the highest level of stroke care within the auspices of larger ‘stroke systems of care’, as advocated by the American Stroke Association.31

A Center for Disease Control and Prevention evaluation of data from the 2005–2007 Paul Coverdell National Acute Stroke Registry recommends that coordinated stroke programs and surveillance of performance measures are needed in order to improve adherence to measures that improve the stroke quality of care.32 The ‘Get With The Guidelines’ stroke program of the AHA is a national stroke quality improvement program developed to reduce disparities in acute stroke care. Voluntary cooperation with this program is associated with substantial and sustained improvement in hospital based acute stroke care regardless of hospital size, geography and teaching status.33 Since the original publication of Primary Stroke Center recommendations in 2000, there have been over 800 facilities certified by the Joint Commission and improved outcomes have been validated by several studies evaluating the stroke systems of care.34

Intra-arterial thrombolysis

Intra-arterial thrombolysis is indicated for the treatment of selected patients with major stroke of <6 h duration due to an occlusion of the middle cerebral artery (AHA Class I, Level of Evidence B, CEBM level 2b, grade B).

Intra-arterial thrombolysis is reasonable for patients who have contraindications to the use of intravenous thrombolysis, such as recent surgery (AHA Class IIa, Level of Evidence C, CEBM level 4, grade C).

IA thrombolysis provides a supplement or alternative to IV thrombolysis in carefully selected patients with AIS. Delivery of fibrinolytic agents directly into the thrombus permits administration of a smaller dose of the agent, which theoretically decreases the risk of cerebral and systemic hemorrhagic complications from systemic effects.35 This concept led to studies that were designed to explore an extension of the treatment window beyond the relatively short therapeutic window for IV rt-PA. Two randomized, multicenter, controlled trials of IA thrombolysis in acute middle cerebral artery stroke have been completed—the Prolyse in Acute Cerebral Thromboembolism Trial I (PROACT-I)18 and PROACT-II.19 Recanalization efficacy and safety of IA recombinant pro-urokinase (r-proUK) for middle cerebral artery occlusion of 6 h duration was demonstrated in the PROACT-I trial. PROACT-II demonstrated a 15% absolute increase in good outcome. The results of PROACT-II, although encouraging, were insufficient to secure FDA approval of r-proUK for an acute stroke indication. For this reason, IA thrombolysis using any thrombolytic agent still represents an ‘off-label’ use of r-proUK or rt-PA. The clinical benefits observed in the PROACT II study were confirmed in a recent meta-analysis of five randomized IA therapy trials, which included PROACT I, and the prematurely halted Japanese Middle Cerebral Artery Embolism Local Fibrinolytic Intervention Trial (MELT) trial, which also found higher rates of good and excellent clinical outcomes in patients undergoing IA therapy.36 In addition, several studies have suggested that successful recanalization correlates with improved clinical outcomes in patients with AIS secondary to large vessel occlusion.37 38 Retrospective studies have shown that IA thrombolysis within 3 h is possible at experienced stroke centers and does not significantly increase risk in patients who have contraindications to IV thrombolysis, such as recent surgery.39–41 The AHA/American Stroke Association Council endorses IA thrombolysis as an acceptable stroke therapy in major stroke of <6 h duration due to an occlusion of the middle cerebral artery in patients ineligible for IV thrombolysis.1 2

Combination treatment

Combination intravenous/intra-arterial therapy is reasonable in selected patients who present with major stroke of <4.5 h duration (AHA Class IIa, Level of Evidence B, CEBM level 2a, grade B).

IV thrombolysis has the advantage of ease and efficiency of administration while IA thrombolysis has the potential for superior recanalization, particularly in proximal occlusions. Combining the two techniques was studied in the 1999 Emergency Management of Stroke (EMS) Bridging Trial42 and the Interventional Management of Stroke (IMS) studies I and II, published in 2004 and 2007, respectively.43 44 The results of IMS II confirmed the safety and increased recanalization rates initially observed in the EMS bridging trial. In IMS I, patients were administered two-thirds dose IV rt-PA followed by IA rt-PA administration. This approach yielded favorable outcomes in 33% of subjects which was significantly greater than the outcome of the historical control group (the placebo arm from the National Institute of Neurological Disorders and Stroke (NINDS) trial) but not significantly different from the historical purely IV thrombolysis group. Both IMS I and II found that IV therapy alone rarely achieved recanalization of large vessel occlusion. Several case series have supported these findings.45–47 Currently, a larger phase III, randomized, controlled IMS III trial is comparing the combined IV/IA approach to standard IV rt-PA therapy alone. Due to the intervening development and FDA clearance of new devices, IA treatment in IMS III allows the use of the EKOS ultrasound microcatheter (which was used to augment thrombus fragmentation in IMS II), the Penumbra thrombo-aspiration system as well as the Concentric MERCI clot retriever device. A recent meta-analysis investigated the use of IA thrombolysis after full dose IV tPA (0.9 mg/kg) and suggested that this method of combined treatment is safe and associated with higher recanalization rates and better functional outcome at 3 months.48

Endovascular thrombus removal

Intra-arterial thrombus removal with the Penumbra aspiration system or Concentric MERCI device is reasonable in selected patients with major stroke where care has been initiated at <8 h duration although data regarding improvement of clinical outcomes is unclear (AHA Class IIa, Level of Evidence B, CEBM level 2a, grade B).

Rapid clot extraction enabling prompt cerebral reperfusion has obvious appeal as a technique for ischemic stroke treatment. This treatment is theoretically ideal for those occlusions of cardioembolic origin, with well organized platelet poor thrombus and extensive fibrin cross linking that is refractory to chemical lysis. The two FDA cleared devices for endovascular clot removal are the MERCI Retrieval System (Concentric Medical Inc, Mountain View, California, USA) and the Penumbra stroke system (Penumbra Inc, Alameda, California, USA). The MERCI device was specifically designed and tested for extracting clots from major cerebral arteries, such as the distal internal carotid artery, the M1 segment of the middle cerebral and the vertebrobasilar arteries. Clots located in more distal cerebral arteries, like the posterior cerebral arteries, distal segments of the middle cerebral arteries or the anterior cerebral arteries are usually not accessible with the MERCI device and the risks of aggressive distal cerebrovascular instrumentation may not be justified by the potential clinical benefit. The safety and efficacy of the MERCI retrieval device for thrombus retrieval was tested in two prospective, single arm, multicenter trials conducted in the USA and led to FDA approval for clot retrieval in the presence of acute stroke.49 However, the approval has been criticized by stroke specialists because neither of the two conducted clinical trials evaluating the MERCI retrieval device were designed to prove its clinical efficacy for stroke therapy.50 Similarly, the Penumbra pivotal trial was a prospective, single cohort, phase 2 study to evaluate the safety and effectiveness of the Penumbra system aspiration catheter in 125 patients with moderate to severe acute stroke due to large vessel occlusion. The FDA approved Penumbra system is an aspiration catheter with a clot separator wire. While the study results showed an unusually high rate of recanalization, only 25% of patients had a good 90 day clinical outcome as defined by a modified Rankin Scale score of ≤2, an outcome that is no better than the natural history of untreated acute stroke patients.21 Subsequent pooled analysis of the MERCI data, however, found that target vessel recanalization was the strongest predictor of outcomes,51 which corroborates findings in IMS I and II.

Other endovascular approaches

The usefulness of other endovascular devices is not yet established, but they may be beneficial and may be considered (AHA Class IIb, Level of Evidence C, CEBM level 4, grade C.)

Angioplasty alone and stent assisted angioplasty have been described, in retrospective series, for the past decade in the treatment of acute cerebral stroke, more recently using existing self-expanding stents approved for stent assisted coil embolization and intracranial atherosclerosis.52–55 Despite a relatively high technical success rate, there are potential immediate and delayed complications of permanent stent placement in acute stroke. The additional risk of hemorrhagic stroke transformation with ongoing antiplatelet therapy is not well defined. Likewise, the incidence of in-stent restenosis, which affects at least 30% of cerebral arteries stented for intracranial atherosclerosis, is not known for intracranial stents implanted for recanalization of acute thrombus.56 A single arm, prospective, FDA approved trial, Stent Assisted Recanalization in Acute Stroke (SARIS), reported data in 20 patients that supports the relative safety and angiographic efficacy of primary stenting in acute stroke.57 Closed cell stent designs allow for resheathing of self-expanding stents, after partial or even full deployment, resulting in a retrievable stent that functions to restore flow temporarily.58 While this may represent a new direction for acute stroke intervention, the concurrent use of anticoagulant, antiaggregating and thrombolytic medications requires further investigation before the routine implementation of this treatment strategy.

Posterior circulation

The usefulness of endovascular treatment in the posterior circulation is not yet established, but it may be beneficial and may be considered, even beyond the 6–8 h time window typical for anterior circulation stroke (AHA Class IIB, Level of Evidence C).

Recanalization rates and outcomes in basilar artery occlusion vary widely for both IA and IV thrombolysis, which is likely due to the wide clinical presentation and differing mechanisms and severity of underlying disease.59 Although occlusion of the basilar artery accounts for only 6–10% of large vessel strokes in humans, several meta-analyses have found that failure to recanalize an occluded basilar artery almost universally results in a poor clinical outcome.59 60 Failure of recanalization, coma at presentation, as well as a proximal location of thrombus have been found to be associated with higher mortality rates.61 Recanalization rates of over 50% have been achieved with both IA and IV thrombolysis in meta-analyses of multiple case series.59 62 A single large prospective registry, the Basilar Artery International Cooperation Study (BASICS), found that in an analysis of 592 patients, 38% of patients were treated >7 h after ictus either with antithrombotics, IV thrombolysis or IA therapy. In this cohort of patients, endovascular therapy did not show superiority over IV thrombolysis in terms of mortality or dependency, after adjusting for multiple factors such as age, stroke severity score and location of the occlusion.63 In the clinical trials of the MERCI and Penumbra thrombus removal devices, posterior circulation occlusions were included and treated with the same time parameters as the anterior circulation occlusion (up to 8 h following ictus) and were found to have similar recanalization rates.20 21 Due to the relatively rare nature of basilar artery occlusion, the poor prognosis and the paucity of prospective data, the rationale for aggressive treatment has been based largely on expert opinion, case series and meta-analyses.64

Adjunctive therapy

Systemic anticoagulation with heparin reduces the risk of catheter related embolism and augments the thrombolytic effect of some agents such as the now infrequently used r-proUK. Another rationale for antithrombotic therapy is prevention of acute re-occlusion. These indications are counterbalanced by the potentially increased risk of brain hemorrhage when heparin is combined with a thrombolytic agent. There is no standard heparin regimen established for IA thrombolysis in acute stroke. Data from the MultiMERCI trial indicates that rates of hemorrhage were not increased when heparin was used.65 PROACT-I18 reported a 27% rate of symptomatic brain hemorrhage with a heparin regimen of 100 U/kg bolus and 1000 U/h for 4 h used with IA r-proUK. Subsequently, a low dose heparin regimen was used in the PROACT studies (2000 U bolus and 500 U/h for 4 h), which reduced the symptomatic brain hemorrhage rate to 7% in PROACT-I and to 10% in PROACT-II.19 GP IIb/IIIa inhibitors have never been studied in a randomized trial of IA thrombolysis in acute stroke but have been used anecdotally to treat re-occlusion and may be necessary during stent placement in the treatment of acute stroke, although dosing and route of administration are variable. A randomized, dose escalation study of IV abciximab in AIS (not exceeding approved doses for coronary intervention) suggested relative safety up to 24 h after onset, although there were non-dose related asymptomatic intracranial hemorrhages.66 Use of the IIb/IIIa inhibitor abciximab for the IV treatment of acute stroke did not demonstrate either safety or efficacy compared with placebo in the phase III trial, Abciximab in Emergency Treatment of Stroke Trial (AbESTT-II).67

Thrombolytic agents

All thrombolytic agents are plasminogen activators and convert plasminogen to its active agent plasmin. rt-PA (alteplase) and r-proUK are the best characterized in the treatment of AIS.18 19 23 Newer agents such as tenecteplase, reteplase, plasmin and microplasmin, and combination therapies to improve efficacy of clot lysis (fibrinolytics and GP IIb/IIIa agents, and fibrinolytics and direct thrombin inhibitors) remain investigational.68–70 Desmoteplase, a fibrin selective plasminogen activator, initially thought to have promise as an IV thrombolytic, showed no benefit over placebo in a phase III trial targeting an imaging defined ischemic penumbra.71 Although some drugs may have differences in systemic effects and hemorrhagic complications, it remains unknown whether one thrombolytic agent is superior to another in terms of revascularization of AIS. Only IV rt-PA is FDA approved for the treatment of acute stroke and all thrombolytics are off-label when administered IA.

Imaging

Imaging recommendations for treatment of acute stroke are included in the AHA guidelines for the early management of adults with ischemic stroke.1 Imaging is a class I recommendation and in most cases a non-contrast head CT will provide sufficient information to guide emergent management decisions; the initial imaging evaluation is required to exclude the presence of hemorrhage or other mass lesion mimicking ischemic stroke.72 There is no Class I evidence not to treat based on other CT findings,1 although some findings have been associated with poorer outcomes, such as the Alberta Stroke Program Early CT Score (ASPECTS), which has been shown to predict outcome in thrombolysis treated stroke patients with signs of acute ischemia on CT.73 The acquisition of vascular imaging, whether CT angiography or MR angiography images with concomitant perfusion imaging during acute stroke, has been shown to be achievable in a reasonable time frame (<15 min imaging time for MRI; 10 min for CT).74 75 CT angiography and MR angiography can be useful to triage patients in terms of large vessel versus small vessel disease and, therefore, to amend treatment options. For this reason, the availability of multimodality vascular imaging in acute stroke has been suggested as a quality metric for a CSC.76 77 Using CT or MR perfusion to identify patients with a mismatch of infarct core and ischemic penumbra, in theory, may allow improved patient selection and/or extension of current treatment time windows and is currently an intense area of research.

The Diffusion and Perfusion Imaging Evaluation for Understanding Stroke Evolution (DEFUSE) investigators noted significant increased odds of achieving a favorable clinical response in patients with MRI defined perfusion/diffusion mismatch treated with IV thrombolysis in the 3–6 h window.78 However, Desmoteplase in Acute Ischemic Stroke Trial II (DIAS-II), the only randomized trial using an ischemic penumbra defined by perfusion imaging to select patients for IV thrombolysis, failed to demonstrate benefit of desmoteplase over placebo in the 3–9 h window.71

In 2010, a technology assessment subcommittee of the American Academy of Neurology concluded that there was insufficient evidence to support the routine use of perfusion imaging in the triage of acute stroke patients in the diagnosis of acute stroke.79 Current evidence states that diffusion weighted imaging is superior to CT for the diagnosis of AIS and indicates that, in anterior circulation stroke, the initial infarct core volume is the best predictor of response to acute stroke therapy.79 Studies designed to determine defining thresholds for futile revascularization and poor outcome have indicated that an ASPECT score of ≤7 and a diffusion weighted imaging core infarct volume of ≥70–100 ml are more frequently associated with poor clinical results.73 80–83