Article Text

Download PDFPDF

Stent retriever technology: concept, application and initial results
  1. S Rohde1,
  2. J Bösel2,
  3. W Hacke2,
  4. M Bendszus1
  1. 1Department of Neuroradiology, University of Heidelberg Medical Centre, Heidelberg, Germany
  2. 2Department of Neurology, University of Heidelberg Medical Centre, Heidelberg, Germany
  1. Correspondence to Dr S Rohde, Department of Neuroradiology, University of Heidelberg Medical Centre, Im Neuenheimer Feld 400, 69120 Heidelberg, Germany; stefan.rohde{at}


Stent retrievers are increasingly used for flow restoration and thrombectomy in acute embolic stroke. First clinical results support the potential of these new devices, in particular the ability to rapidly restore flow and effectively retrieve clots from large intracranial arteries, with favorable clinical results in preliminary patient series. This article reviews the concept and technical aspects of this new technique of endovascular stroke treatment and summarizes the first clinical results.

Statistics from


Ischemic stroke is a leading cause of disability and death in the western hemisphere. In the USA, approximately 795 000 patients experience a stroke per year.1 Systemic thrombolysis with tissue plasminogen activator has proven to be an effective treatment when started early after the onset of clinical symptoms, within a 4.5 h time frame.2–5 However, in severe stroke with thromboembolic occlusion of major cerebral vessels, intravenous lysis is effective in less than 30% of patients.2 ,6 As early recanalization of the occluded vessel is known to be a key factor for treatment success, many efforts have been made to improve the fate of patients with main stem occlusions using a wide range of endovascular maneuvers, comprising local thrombolysis, clot fragmentation and aspiration techniques as well as thrombectomy. Recently, so called ‘stent retrievers’ have been introduced for mechanical thrombectomy in patients with acute embolic stroke, reporting promising results with regard to recanalization rates and safety. This paper reviews the technical aspects of this new technique, and outlines initial clinical results in the context of previous device trials in ischemic stroke treatment.

Evolvement of endovascular recanalization techniques in acute stroke

In the late 1990s, the Prolyse in Acute Cerebral Thromboembolism (PROACT) trials proved the safety and efficacy of intra-arterial thrombolysis with pro-urokinase in patients with acute occlusion of the middle cerebral artery (MCA) up to 6 h after onset of neurological symptoms.7 ,8 In PROACT II, 40% of patients that received interventional treatment had a favorable clinical outcome (modified Ranking Scale (mRS) score ≤2) at the 90 day follow-up, compared with 25% of the control group who received heparin only. There were 10% symptomatic intracranial hemorrhages (sICH) after intra-arterial lysis compared with 2% in the control group. The recanalization rate in the endovascular arm was 66%; however, to demonstrate the pharmacological effect of intra-arterial lysis and to standardize the delivery technique, mechanical clot manipulation was not approved in these trials.

Since then, a variety of mechanical methods have been reported to further improve endovascular recanalization rates—for example, using microwire manipulations, snare devices, balloon angioplasty, thrombus extraction, fragmentation and/or aspiration.9–12 A first approach that was seen as a transition from intra-arterial lysis to mechanical thrombectomy was the application of the EKOS device, adding low intensity ultrasound to enhance the effect of the lytic agent. Intra-arterial treatment in combination with EKOS was tested in a phase I trial in 14 patients13 and later in the Intervention Management of Stroke II (IMS II) trial in 81 patients, resulting in recanalization rates of 57% and 60%, as well as good outcome after 3 months (43% and 46%, respectively).14

In 2005, the MERCI retrieval system (Concentric Medical, Mountain View, CA, USA) was the first device to be approved for intracranial thrombectomy.15 ,16 Basically, the device consists of a corkscrew shaped nitinol coil that engages the thrombus during thrombectomy. Treatment of patients with large vessel occlusions of the anterior and posterior circulation within an 8 h time frame yielded a 57% recanalization rate after mechanical thrombectomy alone. Results could be improved to 69.5% when the device was combined with intra-arterial lysis.16 Symptomatic bleeding occurred in 9.8% of patients.

As a second device, the Penumbra system received CE mark approval.17 The device was specifically designed for vessel recanalization in acute stroke using a debulking and aspiration technique followed by direct thrombus extraction if clot remains. In the Penumbra Pivotal trial, 81.6% of the treated vessels were successfully recanalized to a Thrombolysis in Myocardial Infarction score of 2 or 3 within an 8 h time frame from symptom onset. Procedural adverse events were reported in 12.8%, comprising vasospasm, re-occlusion and vessel perforation. Despite the relatively high recanalization rate, only 25% of the patients achieved an mRS score of ≤2 at the 90 day follow-up. Symptomatic intracranial hemorrhage was reported in 11.2% of patients and all cause mortality was 32.8%. Of note, patients in the Multi MERCI trial had received intravenous lysis before and thus were treated in a ‘bridging’ concept, while a great part of the population in the Penumbra Pivotal trial had intravenous lysis contraindications and thus represent a concept closer to primary mechanical treatment.

For the posterior circulation, numerous case reports and two prospective case series have also reported on a stepwise combined intravenous and endovascular approach. Roth et al tested the Penumbra device after intravenous lysis in 12 patients with basilar artery occlusion and reported a 75% recanalization rate and 66% short term National Institutes of Health Stroke Scale score (NIHSS) improvement.18 Pfefferkorn et al employed a staged approach (comprising intravenous lysis and additional endovascular techniques) in 16 patients with basilar artery occlusion and described remarkably favorable results, with a 94% recanalization rate and 44% good outcome at 3 months.19

Self-expanding stents

The potential benefit of (self-expanding) stents for the treatment of acute stroke was first discussed in 2005 when evaluations of a multimodal recanalization approach found intracranial stent placement to have a major impact on early and successful recanalization.20–22 Theoretically, stent implantation allows almost immediate restoration of blood flow by thrombus entrapment between the stent struts and the vessel wall. Initial clinical results demonstrated technical feasibility and suggested that a self-expanding stent may be highly efficient for recanalization of acute artery occlusion.23 Levy et al reported a recanalization rate of 100% in 20 acute stroke patients that where treated with self-expanding stent systems (Wingspan, n=17; Enterprise, n=2; solely percutaneous transluminal angioplasty, n=124). Intracranial hemorrhage was found in 15% of patients with one sICH (5%). Similar results were reported by Mocco et al who treated patients with acute stroke with the Enterprise stent after routine interventions had been unsuccessful.25 Revascularization was achieved in all patients (75% with a Thrombolysis in Myocardial Infarction score of 3); hemorrhages occurred in 25% and 10% of these bleedings were symptomatic. Despite these encouraging results, a major concern of permanent stent implantation remained the need for subsequent antiaggregation with an increased risk of intracranial hemorrhage and the potential risk of restenosis.22

Stent retrievers


Theoretically, stent placement in acute embolic stroke enables entrapment of the thrombus between the stent and the vessel wall to provide fast recanalization and to restore antegrade blood flow. Flow restoration may rapidly re-establish oxygen supply in the ischemic brain region and enhance the efficacy of thrombolytic drugs. In addition to conventional stenting, stent retrievers allow thrombectomy to be performed by pulling back the deployed stent into the guide catheter, whereby the struts of the stent engage the thrombotic material. Particularly advantageous is that the stent is applicable repeatedly and can be used even in small peripheral vessel branches (eg, M2 segments). In contrast with conventional stent systems, stent retrievers require no anticoagulation or antiaggregation treatment as the stent is not deployed permanently.

Technique of thrombectomy

The intervention is performed under general anesthesia or conscious sedation. A 6–8 Fr guide catheter is placed into the target artery using transfemoral access. To prevent distal thrombus migration and to enhance aspiration during thrombectomy, several groups and manufacturers recommend using a balloon guide catheter for this purpose. A flexible 5–6 F catheter can be used as an intermediate catheter to gain distal access close to the occluded segment. The thrombus is crossed with a 0.021 inch or a 0.027 inch microcatheter. Angiographic runs should be performed through the microcatheter to document the proper position of the microcatheter tip distal to the thrombus and to estimate the length of the clot. The stent retriever is subsequently released by pulling back the microcatheter while holding the retriever device in place. The stent retriever should cover the entire length of the occlusion in order to achieve flow restoration when the stent opens. The device is slowly retrieved together with the microcatheter under continuous aspiration through the guide catheter or the intermediate catheter. In case a proximal balloon catheter is used, the balloon is temporarily inflated to block antegrade flow. The maneuver can be repeated several times to achieve complete clot removal.

An overview of the technical details of different stent retriever devices is given in table 1.

Table 1

Technical features of different stent retriever models

First clinical results

To date, few data are available on the clinical results of stent retrievers; a selection of the published case series is given in table 2.

Table 2

Recent publications about the clinical results of stent retrievers in acute stroke in comparison with the results of former trials investigating intra-arterial lysis (PROACT II) and mechanical recanalization with the MERCI and Penumbra devices

The majority of reports are from applications of the Solitaire (Ev3, Plymouth, MN, USA) stent that was originally developed for the stent assisted treatment of intracranial aneurysms and achieved CE mark for flow restoration in 2009.

As one of the first reported studies, Castano et al assessed the Solitaire in a series of 20 consecutive stroke patients in the anterior circulation within 8 h of symptom onset.26 Thrombectomy was used as rescue therapy in two patients who did not respond to intra-arterial lysis and in three patients in whom successful recanalization with the MERCI retriever was not achieved, resulting in a 90% recanalization rate and good outcome after 3 months in 45% of cases. The mean number of passes was 1.4, and the median time from groin puncture to recanalization was 50 min (range 38–71). Similar results were reported by Roth et al who used the Solitaire stent to treat 22 patients with acute occlusions of the posterior and anterior circulation. There was immediate flow restoration in 21 of 22 cases after deployment of the device, and successful revascularization in 20 of 22 patients. At the 90 day follow-up, 50% showed an mRS score of ≤2; mortality was 18.1% and sICH occurred in 9.9% of the cohort. Our own group employed the Solitaire stent in 18 patients with severe stroke and achieved nearly 90% recanalization.28 Again, successful recanalization could be achieved relatively quickly after the beginning of the intervention (mean 48.3±21.9 min) and the mean number of passes was 2.5. At discharge, a favorable clinical outcome was documented in 33% of patients. Comparable results have been reported for the Solitaire stent as the exclusive device or in combination with intravenous and intra-arterial lysis, or as rescue treatment after unsuccessful use of other mechanical devices in case series with up to 50 patients.29–31

More recently, a case series on a novel, non-detachable stent retriever (Revive, Codman Endovascular) in nine internal carotid artery/MCA and one basilar occlusion reported successful recanalization in all cases and relevant NIHSS improvement at 1 month in 60% of cases.32 Median time to flow restoration and complete recanalization was 27 and 63 min, respectively; mean number of passes was 3.

Currently, there are several multicenter observational studies investigating the safety and efficacy of the Solitaire, Revive and Trevo stent retrievers in acute stroke in larger populations.


Good outcome after acute ischemic stroke depends on many factors but early revascularization plays a major, if not the most important, role among these.6 Intravenous lysis with recombinant tissue plasminogen activator is the currently approved standard treatment with which new stroke treatments have to be compared. However, evidence exists that intravenous plasminogen activation is less efficient in occlusion of proximal large vessels such as the distal internal carotid artery or the main stem of the MCA, and in fact results in less than 30% recanalization in these cases.2 ,6 Although best quality evidence with regard to endovascular treatment comes from the PROACT trials,7 ,8 the lytic (pro-urokinase) used is not commercially available and the focus of development has clearly moved from intra-arterial thrombolytics to thrombectomy devices, especially stent retrievers. The recanalization rates reported in trials, studies and series on endovascular intervention range from 47% to 100% and are thus much higher than those reported for intravenous treatment only.

The above mentioned series on the use of stent retrievers for acute stroke contain some heterogeneity with regard to the respective treatment regimen employed before stent application (ie, intravenous lysis in many cases) and to treatment time point (<3 to < 8 h from stroke onset). The reported rate of sICH ranges from 0% to 20% with an average of around 10% and is therefore comparable with the sICH rate of former trials using endovascular recanalization techniques. However, the cited studies clearly suggest a potential clinical benefit of stent retrievers by rapidly, safely and effectively retrieving clots from large intracranial arteries. Compared with former mechanical devices, stent retrievers have the advantage of easy handling and almost immediate restoration of antegrade flow after deployment of the device, independently of successful thrombectomy, thus potentially saving time for the interventionalist and the brain. Moreover, the clot is not disrupted mechanically but extracted entirely. Regarding procedural safety, device related complications, vasospasm or thrombus fragmentation have been reported only in rare cases.29

However, with regard to the impressive recanalization rates, clinical outcome remains to be improved as the percentage of patients with an mRS score <2 at 3 months rarely exceeds 50%, the rate of sICH averaging about 10% and mortality ranging between 26% and 36%. This discrepancy between good technical and still less satisfactory clinical results might have several causes, such as severity of stroke, patient selection, time to treatment, duration of procedure or concomitant anticoagulation treatment.

To address one of these factors, time to treatment in many reports was beyond the 4.5 h window, as patients were included for up to 6 h or 8 h, and at times even up to 12 h. Initiating mechanical treatment earlier, together with an improved imaging based patient selection (still to be established) should increase rates of good outcome. This and other factors thus need to be further evaluated and improved before mechanical endovascular therapy in general and stent retrieval in particular can be seen as a routine treatment of acute ischemic stroke.

Above all, a randomized, multicenter trial on mechanical endovascular stroke therapy is warranted to eventually leave the evidence level of non-controlled single arm trials and case series. This trial should address the question of whether the outcome in main trunk occlusion with the approved standard therapy of intravenous lysis can be improved by additional mechanical thrombectomy using stent retrievers.



  • Competing interests None.

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

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.