During the initial few years, the technique of stent-assisted revascularization to manage acute ischemic stroke was directly adopted from cardiology with little modification. Subsequently, with growing experience and better understanding, the stent technology was adopted with some amendments and finally led to the development of dedicated neurovascular stents. In the following section, we will discuss the evolution of stent technology beginning with balloon-mounted stent moving on to self-expanding stents and then to stent retrievers.

Balloon-mounted cardiology stents were first used for revascularization in acute ischemic stroke. The stent was mounted on a balloon which could be expanded to the desired size. Despite the initial encouraging results, it was soon realized that the balloon-mounted stents were not optimal for cerebral vasculature. The degree and pressure of the balloon deployment were the main concerns. Often the lack of adequate information on the vessel diameter can lead to inappropriate over inflation of the balloon with the risk of vascular injury or underinflation resulting in incomplete stent deployment. The pressure of the balloon inflation could lead to arterial rupture or dissection irrespective of the degree of inflation. Other factors like deliverability, trackability, and conformability of the balloon-mounted stent system in the often tortuous cerebral vasculature posed technical challenges and safety concerns.

The next phase started with the development and use of self-expanding stents (SES). These stents were primarily used during treatment of cerebral aneurysms as well as intracranial stenosis. Many centers used these stents as off-label for revascularization in acute ischemic stroke. The self-expanding stents are sheathed and designed to self-expand to a predetermined size when unsheathed. The ability to self-expand obviated the need for a balloon-assisted expansion and the resulting vascular complications. A microcatheter along with the enclosed stent is positioned just beyond the thrombus location. The stent is then slowly unsheathed allowing for the stent to self-expand and compress the thrombus against the vessel wall resulting in recanalization. Some of the self-expanding stents include Wingspan (Stryker, Natick, MA, USA), Enterprise (Cordis Neurovascular, Miami Lakes, FL, USA), and Neuroform system (Stryker, Natick, MA, USA).

Though the balloon-mounted stents and the self-expanding stents were never FDA approved for mechanical thrombectomy in acute ischemic stroke, they were used off-label for this purpose. Also, both the stents had to be left behind in the cerebral vasculature which needed immediate and long-term antiplatelets to prevent acute as well as in situ stent thrombosis later on. Though this is a standard of care in cardiology with a high safety margin, the use of antithrombotics (antiplatelets/GpIIb/IIIa antagonists) is risky in AIS because of the additional risk of intracerebral hemorrhage when used in conjunction with IV tPA. Also, hemorrhagic transformation of infarct can lead to a worse prognosis in the presence of antithrombotics. The big breakthrough happened when the retrievable stents were introduced which will be discussed below.

An entirely retrievable stent was initially developed as a revascularization device to be used during aneurysmal coiling in 2003. Though the potential utility of these new stent retrievers in mechanical thrombectomy was hinted at, it was not until a few years later when the first successful reports were published. Coupled with suboptimal results using the then approved mechanical thrombectomy devices and the need for more fast and efficient revascularization, the stent retrievers took the center stage for experimentation in mechanical thrombectomy for ischemic stroke opening up an entirely exciting field of stroke therapeutics which led to the innovation and further development of the stent retriever technology and multiple multicenter randomized clinical trials culminating in successful demonstration of an effective endovascular stroke therapy for AIS.

These stent retrievers are similar to the self-expanding stents described earlier with regard to unsheathing and deployment (self-expanding); in addition, they remain attached to a wire at their proximal end and which in some cases could be electrolytically detached. As of 2016, some of the stent retrievers available in the market includeSolitaire FR (Covidien/Medtronic), Trevo XP (Stryker Neurovascular), REVIVE SE (Codman Neurovascular), and pREset & pREset Lite (Phenox). Though all of them have a similar physical shape, they differ from one another in the cell design, radial force exerted by the stent, orientation of the struts, radiopaqueness, size available, detachability, etc. A detailed description of the individual stents and their properties is beyond the scope of the current chapter, and a few salient features of the stents are provided below (Table 8.2). In Fig. 8.1, an image of the Trevo stent retriever is shown demonstrating its use.

The first reported use of the Solitaire AB (aneurysmal bridging) stent retriever for ischemic stroke was in 2009 when it was successfully used for mechanical thrombectomy of a middle cerebral artery M1 segment occlusion after a failed attempt with the MERCI retriever. This was soon followed by multiple case series (involving approximately 280 patients) reporting the use of Solitaire AB stent retriever for mechanical thrombectomy. In these case series, the Solitaire stent retriever was used both as a primary device and the rescue treatment when other modalities like IV tPA and/or mechanical thrombectomy using other devices failed.

Simultaneously, another stent retriever, Trevo (Stryker Neurovascular), was developed for the treatment of mechanical thrombectomy in acute ischemic stroke. After initial animal studies which showed safety and efficacy, a small case series and the Trevo study showed the feasibility, safety, and utility of the Trevo stent retriever in mechanical thrombectomy of acute ischemic stroke.

These initial encouraging results, the suboptimal performance of the then existent mechanical thrombectomy devices, led to the first-generation randomized clinical trials of the stent retrievers against the MERCI concentric retriever—SWIFT and TREVO2 which will be discussed below.

SWIFT was a randomized, parallel group non-inferiority trial in which 113 patients with moderate to severe ischemic stroke were randomized to the Solitaire FR (flow restoration) stent retriever (n = 58) or MERCI retriever (n = 55) [1]. The trial recruitment was halted by DSMB after an interim efficacy analysis during which the pre-specified criteria for trial termination were met. The primary end point of successful recanalization (with the assigned study device and not needing rescue treatment) without symptomatic intracranial hemorrhage occurred in 61% with Solitaire FR stent retriever compared to 24% with MERCI retriever. Core lab-assessed rates of angiographic recanalization with Solitaire FR stent retriever was 69% compared to 30% with the MERCI retriever. The need for rescue treatment was also lower in the Solitaire FR group (21%) compared to the MERCI retriever (44%). Good neurological outcome at 90 days (modified Rankin scale of 2 or less, NIHSS score improvement of 10 points) was seen in 58% with Solitaire FR stent retriever compared to 33% with MERCI retriever. The safety end points (study device related, symptomatic ICH, death from any cause at 90 days) were also lower in the Solitaire FR group (9%, 2%, 17%, respectively) compared to the MERCI retriever (16%, 11%, 38%, respectively).

TREVO2 was also an open-label randomized non-inferiority trial in which 178 patients were randomized to the Trevo retriever (n = 88) or MERCI retriever (n = 90) [2]. The primary efficacy end point of TICI ≥ 2 reperfusion (using assigned device—as determined by core lab) was 86% with the Trevo retriever compared to 60% with the MERCI retriever. There was a non-statistically significant difference in the need for rescue therapy in the Trevo retriever compared to the MERCI retriever (18% vs. 31%, p = 0.0851). Good clinical outcome at 90 days (mRS 0–2) is seen in 40% with Trevo retriever and 22% with MERCI retriever. Despite meeting some of the efficacy end point, the composite safety primary end point and mortality at 90 days were not statistically different between Trevo retriever and MERCI retriever (15% vs. 23%, p = 0.18 and 22% vs. 24%, p = 0.18, respectively).

Following these trails, FDA approved both the stent retrievers for mechanical thrombectomy in acute ischemic stroke in 2012. The three major randomized trials (IMS III, SYNTHESIS, and MR RESCUE) that were finished in 2012 failed to show the expected superiority of the endovascular stroke therapy compared to intravenous thrombolysis with tPA alone. In addition to other reasons, the use of older generation devices with their drawbacks and very limited use of these novel stent retrievers in these trials contributed to the negative trial results. With a better understanding of the reasons for failure and availability of technologically advanced mechanical thrombectomy devices, researchers around the world embarked on multiple randomized controlled trials pitting the endovascular stroke therapy (predominantly using the novel stent retrievers) against standard IV thrombolysis using tPA which will be discussed below.

The role of stent retrievers in achieving high rates of recanalization with speed and safety has been demonstrated in five recent endovascular trials—MR CLEAN, ESCAPE, SWIFT PRIME, EXTEND-IA, and REVASCAT [3–7]. MR CLEAN is the first trial to have successfully shown the superiority of the endovascular treatment, the results of which were presented at the World Stroke Congress in October 2014. Following the release of the MR CLEAN trial results, two of the other trials ESCAPE and EXTEND-IA temporarily withheld the trial recruitment and planned interim analysis. SWIFT PRIME followed suit after a few days. The trial results of ESCAPE, SWIFT PRIME, and EXTEND-IA were subsequently presented at the International Stroke Conference in February 2015 followed by the release of the results of REVASCAT trial a few months later. The overwhelming positivity of all the five trials, consistent results across trials in various health systems across the globe, led to a long overdue paradigm shift in stroke care. We will discuss individual trials followed by pooled analysis of patient-level data from all the five trials.

Multicenter randomized clinical trial of endovascular treatment for acute ischemic stroke in the Netherlands (MR CLEAN) was a multicenter prospective randomized open-label blinded end point evaluation clinical trial run at 17 centers in the Netherlands comparing intra-arterial treatment versus no intra-arterial treatment for acute ischemic stroke [3]. The trial design was pragmatic and closely resembles the routine clinical practice. The investigators sought out to explore if any form of intra-arterial therapy (IAT) was superior to usual care (control, including the use of IV tPA) in acute ischemic stroke with proximal anterior circulation occlusion. All modalities of intra-arterial treatment (thrombolysis using tPA, urokinase, MERCI retriever, Penumbra aspiration system, sonolysis, stent retrievers) were approved to be used in the trial, and there was no restriction on the use of multiple modalities for a patient in whom one modality fails. Also, the need for performance of an acute cervical carotid stenting during thrombectomy was left to the discretion of the local interventionist. Patients were eligible to be randomized into the trial irrespective of their status with regard to eligibility, contraindications, treatment status, and response to IV tPA. The trial enrolled 500 patients with 233 (46.4%) randomized to the intra-arterial treatment. Despite approval of a broad range of the IA treatment modalities, 81.5% (190/233) of the patients in the trial were treated with stent retrievers. The exclusive use of other modalities like IA thrombolysis (1/233), MERCI retriever (2/233), and thromboaspiration (1/233) was very limited. The use of IV tPA was comparable across the intervention (87.1%) and control groups (90.6%). 90-day median mRS was chosen as the primary outcome for the trial which was lower in the intervention group with an adjusted common odds ratio of 1.67. Shift analysis of the mRS scores also favored the intervention for mRS groups 0–5; there was, however, no mortality benefit between both the groups. All the clinical and imaging secondary outcomes also favored the intervention group. Despite minor limitations of the trial, MR CLEAN trial results closely resemble the complex situations seen in routine clinical practice, thus increasing the generalizability of the trial results. The trial design, baseline demographic characteristics, imaging features and time metrics, and the trial results are presented in Tables 8.3, 8.4, 8.5 and 8.6, respectively.

Endovascular treatment for small core and anterior circulation proximal occlusion with emphasis on minimizing CT to recanalization times (ESCAPE) was a prospective, multicenter randomized, open-label, controlled trial with blinded outcome evaluation [4]. The trial was run worldwide at 22 centers, predominantly in Canada. The trial recruitment was kept on hold after the results of the MR CLEAN trial were released. The data safety monitoring board advised the trial to be stopped due to efficacy after the unplanned interim analysis. At that point, the trial had enrolled 316 patients out of an initial target sample size of 500. The trial compared the endovascular treatment along with the usual standard of care (intervention) with the usual standard of care alone (control) in acute ischemic stroke patients older than 18 years with an NIHSS > 5 and an occlusion in the proximal anterior circulation and small ischemic core and large penumbra. Though the use of any mechanical thrombectomy devices as per local site availability was allowed, the emphasis was placed on the use of retrievable stents by the trial investigators. Similar to the MR CLEAN trial, patients were eligible to participate in the trial irrespective of their eligibility or treatment status with respect to IV tPA. Some of the unique features of the trial included the extended time window, emphasis on patient selection, and workflow during the trial execution. Patients were eligible to be enrolled up to 12 h from symptom onset. There was a trend toward a better outcome in the patients enrolled beyond 6 h from symptom onset. However, the numbers were not sufficient to achieve statistical significance. The trial investigators aimed to select patients with good premorbid function coming in with an acute moderate-severe disabling ischemic stroke with demonstrable occlusion in the proximal anterior circulation with a small ischemic core and ability to initiate rapid endovascular treatment after stroke imaging. To achieve this goal, patients were carefully selected using various clinical (modified Barthel index) and imaging tools (CT ASPECTS scoring, CT angiography, multiphase CT angiography, CT perfusion). A lot of emphases was placed on achieving target metrics during the patient care workflow during the trial recruitment and execution. The investigators adopted target metrics of 60 min for study CT-to-groin puncture and 90 min for study CT-to-first reperfusion to be achieved for the patients enrolled in the trial. Constant weekly monitoring of the quality of the participant recruitment and workflow metrics at individual sites during trial recruitment was done, and appropriate feedback was provided to the individual sites. The investigators emphasized the role of parallel processing of the clinical and imaging information as well as parallel decision making with the goal to achieve the fastest reperfusion by the endovascular route. The above mentioned components of the ESCAPE trial design were crucial to its success. The trial design, baseline demographic characteristics, imaging features and time metrics, and the trial results are presented in Tables 8.3, 8.4, 8.5 and 8.6 respectively.