Endovascular Treatment of Stroke

Background

According to estimates by the World Health Organization (WHO), stroke is the second leading cause of death worldwide and is the third greatest contributor to lost disability-adjusted life years (DALYs) globally. There are approximately 17 million strokes worldwide each year, with a disproportionately high incidence in low-income countries. In the United States, approximately 795,000 patients have a stroke annually, and with a mortality rate of 18%, strokes are the fifth leading cause of death in the country. Overall direct and indirect healthcare costs of stroke are significant, costing US healthcare payers more than $34 billion each year.

A majority of strokes (87%) are ischemic in etiology, rather than hemorrhagic. Such ischemic strokes can be embolic in nature ( due to emboli arising from the heart or peripheral vasculature) or flow-related (due to chronically stenotic internal carotid artery [ICA] or intracranial atherosclerotic disease [ICAD]). The INTERSTROKE study, a large, international case-control study identified several important, potentially modifiable risk factors for stroke, including hypertension, diabetes, lack of physical activity, poor diet, smoking, alcohol use, and waist-to-hip ratio. These risk factors were collectively associated with 90% of all strokes globally.

Traditional Treatment of Ischemic Stroke

Neuronal structures and synaptic connections in the brain are exquisitely sensitive to hypoxia. As a result, regions of the brain affected during an ischemic stroke permanently lose approximately 1.9 million neurons and 14 billion synapses every minute of ischemic time. Therefore, treatments of acute stroke have been focused on expediting brain reperfusion with either pharmacologic or mechanical techniques.

In 1996, the US Food and Drug Administration (FDA) approved the use of intravenous (IV) alteplase (tissue plasminogen activator [tPA]) for patients with acute ischemic stroke presenting up to 3 hours after symptom onset. Based on the positive results of several randomized, controlled trials, tPA has since been routinely used as the standard of care in acute ischemic stroke. In 2010, the time window for tPA was further expanded to 4.5 hours following the results of the ECASS III trial. ^,

The timing effects of fibrinolytic therapy have been firmly established, with earlier time to treatment associated with improved rates of independent outcomes and reduced rates of symptomatic intracranial hemorrhage (ICH). A pooled analysis of tPA clinical trials published in 2004 showed a highly significant association between earlier treatment with tPA and favorable 3-month outcome.

Despite its proven efficacy, treatment of ischemic stroke with tPA certainly has limitations, including incomplete recanalization and risk of hemorrhage. Proximal large vessel occlusions of the ICA have exceedingly low rates of recanalization following administration of tPA (<10%), with rates increasing as the site of occlusion migrates distally (approximately 25% of M1 and 40% of M3 occlusions). ^, Furthermore, early results observed high rates of hemorrhagic conversion of ischemic stroke, in some cases as high as 20% of patients treated with tPA. However, more recent studies have demonstrated an improved safety profile, with symptomatic hemorrhage rates of approximately 3%, potentially due to stricter patient selection criteria.

Intravenous tPA remains the only pharmacologic thrombolytic agent approved for use in acute ischemic stroke, and until the development of endovascular techniques for mechanical thrombectomy, IV fibrinolysis was the only treatment for acute stroke. Even as such, estimates suggest that only 25% of ischemic stroke patients present early enough to be considered candidates for tPA. Thus there emerged a need for improved treatment options for patients with acute ischemic stroke.

Evolution of Endovascular Stroke Therapy

In the 1980s, early case reports began to emerge of patients with acute large vessel occlusion undergoing cerebral angiography and focal intra-arterial fibrinolytic therapy, attempting to mimic the success of coronary interventional revascularization. Further case series suggested the feasibility of the procedure, as well as improved outcomes in patients who achieved sustained revascularization. ^, These results prompted the development of the first endovascular randomized controlled trials for stroke: Prolyse in Acute Cerebral Thromboembolism (PROACT and PROACT-II ). These trials randomized patients with ischemic stroke due to occlusion of the M1 or M2 segment to either intra-arterial fibrinolysis or placebo (PROACT), or intra-arterial fibrinolysis plus heparin or heparin alone (PROACT-II). Both trials observed increased rates of recanalization in the intra-arterial treatment arms, and furthermore, the results of PROACT-II suggested improved 90-day outcomes after intra-arterial therapy. Of note, however, rates of symptomatic ICH were higher in the treatment arms in both trials, and there was no observed change in overall mortality. With the publication of these initial trials came a further diaspora of trials for interventional acute stroke therapy. Early pilot studies published the feasibility and efficacy of combined IV and intra-arterial tPA (Emergency Management of Stroke, EMS ; and Interventional Management of Stroke, IMS). ^,

In 2005, a pivotal paradigm shift occurred with the publication of the prospective MERCI Trial, proving the safety and efficacy of the MERCI intracranial embolectomy device. The initial MERCI device was a corkscrew-shaped coil retriever navigated through an embolus to then engage and retract the clot from the intracranial vasculature. The study investigators observed increased recanalization rates and improved clinical outcomes following mechanical thrombectomy, as well as acceptable complication rates. Results of the Penumbra trial, utilizing the Penumbra suction thrombectomy catheter to debulk and aspirate distal emboli, also demonstrated improved recanalization rates with an acceptable safety profile. Despite initial success as proofs-of-concept, these early thrombectomy devices had technical limitations, and thus several novel devices were subsequently developed in the form of stent-retrievers. These devices, Trevo (Stryker) and Solitaire (Medtronic), improved recanalization rates by engaging thrombi at multiple locations and further allowed for earlier recanalization by expediting flow through the clot as the device expanded. ^, These results directly paved the way for the development of endovascular randomized, controlled trials for large vessel occlusion, causing acute ischemic stroke.

Clinical Trials for Mechanical Thrombectomy

A complete list of pivotal clinical trials in stroke thrombectomy can be reviewed in Table 59.1 .

In 2013, the first three randomized controlled trials for mechanical thrombectomy (SYNTHESIS, IMS-III, MR RESCUE) were published concurrently in the New England Journal of Medicine . Each of these trials failed to demonstrate a difference in clinical outcomes (improvement to mRS ≤2) with mechanical thrombectomy compared to standard medical therapy alone. However, several limitations have been noted with these trials. First, they used an older generation of thrombectomy devices, and because the trials lasted between 5 and 7 years, the investigators were not able to adapt to improvements in technology. Second, to be enrolled in the trials, patients did not require confirmatory imaging of large vessel occlusion. Lastly, treatment times were often delayed. Despite these limitations, several important conclusions were drawn. Importantly, the risks of interventional treatment did not significantly increase rates of hemorrhagic conversion or symptomatic ICH. They also reinforced the concept that the benefits of tPA depend on the site and length of embolic occlusion, with improved outcomes for smaller, more distal clots. ^, Further analysis of the IMS-III cohort demonstrated an association between faster time to intervention and improved neurologic outcomes after thrombectomy.

With these important considerations, a subsequent generation of clinical trials were designed, utilizing newer-generation stent-retrievers. Five rigorously performed trials (MR CLEAN, ESCAPE, EXTEND-IA, REVASCAT, SWIFT PRIME ) demonstrated improvements in clinical outcomes after mechanical thrombectomy for large vessel occlusion, compared to patients receiving standard medical therapy. In fact, interim analysis of the MR CLEAN cohort demonstrated such impressive results that the remaining trials were terminated after similarly impressive outcomes, and two remaining trials (THRACE and THERAPY ) were terminated prematurely. A subsequent meta-analysis of the clinical trials further highlighted these positive outcomes, observing increased rates of angiographic revascularization and functional outcome with thrombectomy compared with medical therapy alone, but no difference in symptomatic ICH rates or overall mortality between treatment arms. The number of patients with large vessel occlusion needed to treat in order to prevent one disability (NNT) was 8, which is comparable to the treatment effects of tPA (NNT = 8) and comprehensive stroke units (NNT = 11).

With these promising results, mechanical thrombectomy for acute large vessel occlusion became the standard of care for appropriately selected patients. As a result, the American Heart Association/American Stroke Association (AHA/ASA) revised their Guidelines for the Early Management of Acute Ischemic Stroke to include mechanical thrombectomy in 2013. The indications for mechanical thrombectomy, based on level 1A evidence, were defined as follows: (1) Prestroke Modified Rankin Scale (mRS; Table 59.2 ) of 0 or 1; (2) causative occlusion of the ICA or middle cerebral artery (MCA) segment M1; (3) age ≥18 years; (4) National Institutes of Health Stroke Scale (NIHSS, Table 59.3 ) of ≥6; (5) Alberta Stroke Program Early CT Score (ASPECTS; Fig. 59.1 ) of ≥6; and (6) treatment can be initiated within 6 hours of symptom onset. Although based on less-rigorous evidence, the guidelines suggest that treatment can also be considered for patients with MCA M2 or M3 segment occlusion, anterior cerebral artery (ACA) occlusion, posterior cerebral artery (PCA) occlusion, or vertebrobasilar occlusion. Furthermore, the guidelines propose a technical goal for thrombectomy to maximize the probability of functional outcome as a Thrombolysis in Cerebral Infarction Scale of 2b or 3 (TICI; Table 59.4 ), defined as complete but sluggish filling of the distal vascular territory or complete revascularization, respectively, and recommend achieving this outcome as early as possible.

Since the benefit of thrombectomy has been definitively established, there is a continued process underway to determine other patient cohorts that may similarly benefit from endovascular treatment of acute ischemic stroke. Most recently, two clinical trials (DAWN and DEFUSE 3 ) highlighted the benefit of thrombectomy in patients presenting outside the recommended treatment window (6 and 24 hours and 6 and 16 hours after symptom onset, respectively), provided there was a viable, salvageable penumbra based on perfusion imaging. These trials were included in the updated 2018 AHA/ASA Management of Acute Ischemic Stroke Guidelines, extending the time to treatment to 24 hours in eligible patients. No doubt, as further evidence emerges, the indications for mechanical thrombectomy will continue to rapidly evolve.