Baker et al. published a systematic review of neurothrombectomy devices for the treatment of acute ischemic stroke in 2011. This review included any human studies that reported on outcomes following thrombectomy. A total of 87 articles met the inclusion criteria, 62 of these were case series or case reports, 18 were prospective single-arm studies and seven were retrospective single-arm studies. The rate of successful recanalization, defined as Thrombolysis in Myocardial Infarction (TIMI) flow grade of two or two, ranged from 43-100% across all studies. Higher rates of recanalization were reported with the Penumbra System (83-100%) compared to either the Merci Retriever (43-78%) or other devices (50-90%). Clinical effectiveness was determined by a post-treatment Rankin score of 0-2, a measure that was available in 17/25 studies. There was a wide range of clinical effectiveness, from 15% to 60% of treated patients. The most common adverse events from treatment were intracranial hemorrhage, and vessel perforation/dissection. The rate of symptomatic intracranial hemorrhage ranged from 0-25%, and the rate of asymptomatic intracranial hemorrhage ranged from 1-43%. Target vessel perforation or dissection was reported by less than half the studies, with rates ranging from 0-7%.
In 2008, Stead and colleagues conducted a systematic review and meta-analysis of percutaneous clot removal devices. Of note, the authors were unable to obtain individual patient data for the Mechanical Embolus Removal in Cerebral Ischemia (Merci) trial described below; those 151 patients were not included in the meta-analysis. The authors identified 14 case series and 8 case reports with a total of 147 patients. The Merci® Retriever was utilized for 17 patients; a variety of mechanical embolectomy devices (with coronary or peripheral vascular indications) were used in other studies. Patients were similar in that they were diagnosed with large vessel disease but were otherwise heterogeneous. Emboli were accessible in 85% of patients. In all studies, post-procedural blood flow was measured using the TIMI grade. A flow rate representing full recanalization was achieved in 67 of 146 patients (45.6%). Partial or full recanalization was achieved in 101 of 146 patients (68.7%). When embolectomy methods were compared, superiority of one device over others was not demonstrated in accessing the lesion, retrieving the clot, or in clinical outcome. Pooled data were compared to the placebo arm and intra-arterial thrombolysis arm of the PROACT II (Prolyse in Acute Cerebral Thromboembolism II) study, comparing intravenous and intra-arterial tPA use. Partial or full
recanalization rate in the placebo group was 18%; the rate was 66% in the intra-arterial group. However, the authors acknowledge that 81 patients (55.1%) in the meta-analysis also received thrombolytics, and the comparative role of thrombolytics against mechanical thrombectomy is unknown. The authors concluded that there was a modest survival benefit in the mechanical thrombectomy patients compared to historical controls, while recognizing the limitation of small study sizes and non-randomized comparator groups.
Two single-arm studies were published in 2011 that made comparisons with historical controls. Alexandrov et al. treated 125 patients presenting with acute stroke with the Penumbra System. Outcomes of embolectomy were compared with historical controls that were treated with IV tPA in a previous clinical trial. Embolectomy patients had a similar stroke severity score, but were younger and had a longer time from onset of treatment to symptoms. The rate of recanalization was 82% for the embolectomy patients; this was higher than the 40% recanalization rate reported with TPA. However, mortality at three months was higher in the embolectomy group compared to TPA (32.8% vs. 14.1%, p=0.008), and the rate of favorable functional outcome was lower (25% vs. 39%, p=0.046).
Taschner et al. treated 22 consecutive patients with acute ischemic stroke and a National Institutes of Health (NIH) stroke scale score of at least seven with the Penumbra system. Outcomes from this group of patients were compared to patients treated with TPA who were matched for stroke score and location. Recanalization with embolectomy was successful in 25/32 target vessels (78%) compared to 17/32 (53%) with TPA. A favorable outcome, defined as a stroke score of 0-1 or an improvement of at least ten points, was present in 2/20 (10%) of patients treated with embolectomy, compared with 7/20 (35%) treated with TPA.
In 2005, Smith and colleagues reported the results of the MERCI trial. This was a multicenter (25 centers), prospective nonrandomized trial of this device for patients with symptoms of acute stroke for less than eight hours who were not candidates for thrombolytic therapy, either because of contraindications (approximately 25%) or because symptoms were present for more than three hours. A total of 1,809 patients were screened to identify the 151 patients enrolled in the trial. Chief reasons for exclusion were NIHSS (National Institutes of Health [NIH] stroke score) too low or improving, intracranial hemorrhage, or inability to obtain consent. Of the 151 patients, 141 had the device deployed. Recanalization was achieved in 46% (69/151) of patients on an intent-to-treat analysis and in 48% (68/141) of patients in whom the device was employed. (One patient had “spontaneous” recanalization.) The status of vessels distal to the treatable vessel was not considered in the recanalization rate. Clinically significant procedural complications occurred in 10 patients (7.1%), and symptomatic intracranial hemorrhages were observed in eleven (7.8%). Good neurological outcomes were more frequent at 90 days in those with successful recanalization compared to those with unsuccessful recanalization (46% vs. 10%, respectively; p<0.0001) and mortality was less (32% vs. 54%, respectively; p=0.01). Of note, in the study, up to six passes could be made to remove the clot, and at least two devices were used in each patient in the MERCI trial. The MERCI investigators compared their patients to the placebo arm of the PROACT II study to determine safety and efficacy of mechanical embolectomy.
In 2008, Smith and colleagues reported the results of the Multi MERCI trial, a prospective, international, multicenter, single-arm study. As with the MERCI trial, patients were eligible if they presented with eight hours of onset of symptoms from large-vessel stroke. In addition to the MERCI indications, patients were eligible if they received intravenous tPA but failed to completely recanalize their occluded vessel. A total of 1,088 patients were screened to enroll 177 patients. Of these, 164 patients had the device deployed. A newer generation device was available for 131 of the 164 patients, and patients could be treated with adjuvant intra-arterial tPA, depending on the operator. Recanalization was achieved in 55% (90/164) on intention-to-treat analysis, and in 58% (88/151) in the per-protocol analysis. Two patients recanalized spontaneously. Procedural complications occurred in nine patients (5.5%), and symptomatic intracranial hemorrhage was observed in 16 (9.8%). In comparison with patients who did not recanalize, 90-day neurological outcomes favored patients in whom flow was restored (49% vs. 10%, respectively; p=0.001). An average of three attempts was made on each patient. This report also compares their results to the placebo arm of the PROACT II trial.
To determine if this treatment improves net outcomes (considers both benefits and risk) in stroke, there must be a comparison with an appropriate control group. It is not clear what the recanalization rate would have been without embolectomy in those patients who had successful clot removal. Concurrent control groups are also important to evaluate possible unexpected events when intravascular devices are used that may damage arterial endothelium.
Concerns have been raised about using the patients from the PROACT II study as historic controls. These concerns include the fact that the MERCI trial included patients with different types of occlusions; PROACT II had M1 and M2 occlusions while the MERCI trial also included internal carotid and vertebral basilar systems. Questions have also been raised about the outcome measure of recanalization since the MERCI study did not look for distal emboli. Also, there are concerns about the reliability of the TIMI perfusion score as reported in this trial, and thus questions about whether the recanalization rates can be compared among studies.
In 2010, Shi and colleagues published a retrospective subgroup analysis of data from the MERCI and Multi MERCI trials, comparing patients with middle cerebral artery occlusions of the first (M1) or second (M2) large vessel branch. The relatively smaller M2 occlusions were associated with higher recanalization rates. Recanalization was also improved in patients who had received intravenous tPA. However, an accompanying editorial points out that no difference in 90-day morbidity or mortality was noted. Molina further comments that in published literature of patients undergoing mechanical embolectomy, recanalization is often achieved without improvement in clinical outcomes at three months post-procedure. Rates of “futile” recanalization range from 26% to 51% in various studies.
In 2007, Flint and colleagues reported on a study of 80 patients who had use of the MERCI device for occlusion of the intracranial internal carotid artery. Forty-seven of these patients were from the MERCI trial, and 33 were from the Multi-MERCI study, which included patients who had been previously treated with intravenous tPA. Of these 80 patients, 53% had internal carotid artery (ICA) recanalization with the MERCI device and 63% had ICA recanalization with the MERCI device and adjunctive treatment. Successful clinical outcomes as defined by the study were more common in study patients who had successful recanalization than in study patients for whom recanalization was not successful. As noted above, this was a single-arm study.
In 2009, Lin and colleagues reported on a case series of 75 patients with internal carotid artery terminus occlusion. The team performed a retrospective analysis of ischemic stroke patients at their institution who received a variety of treatment options, alone or in combination. Primary outcome measure was recanalization rate. Lowest recanalization rate was observed with intra-arterial thrombolytics alone at 17.6% (3/17), while Merci embolectomy with intra-arterial thrombolytics was associated with the highest recanalization rate at 85.7% (18/21). The small sample sizes, as well as non-randomized and retrospective design preclude definitive conclusions from these results.
Published articles identified report small (approximately 25 patients) non-comparative results from single centers: University of California at Los Angeles (UCLA) and University of Tennessee. The published evidence continues to describe case series and often includes only intermediate outcomes, such as vessel recanalization. In addition, results from a trial using the Penumbra System® on 23 enrolled patients (21 target vessels) are available.
An ongoing trial, the Magnetic Resonance and Recanalization of Stroke Clots Using Embolectomy (MR RESCUE) study, being sponsored through the National Institutes of Neurological Disorders and Stroke, is underway. This study randomly assigns patients with stroke seen within eight hours to either medical therapy or embolectomy. This trial should provide important comparative data.
Practice Guidelines and Position Statements
The 2007 guideline for early management of ischemic stroke from the American Heart Association and others gives a class II B recommendation (usefulness/effectiveness is uncertain) to the use of the mechanical embolectomy devices. This guideline notes that the utility of the devices in improving outcomes after stroke is unclear. This position was reiterated in 2009 in a scientific statement from the American Heart Association. In this scientific statement the following recommendations are made:
- Although the Concentric Merci device can be useful for extraction of intra-arterial thrombi in appropriately selected patients, the utility of the device in improving outcomes after stroke remains unclear (Class IIb, Level of Evidence B).
- The usefulness of other endovascular devices is not yet established, but they may be beneficial (Class IIb, Level of Evidence C).
The scientific evidence does not permit conclusions concerning the effect of mechanical embolectomy on patient outcomes. The existing observational data are not sufficient to determine whether this approach improves net health outcomes. Single-arm studies report a high rate of recanalization of the infarcted vessel, however only a subset of patients with successful recanalization achieves good functional outcomes. Comparison with historical controls receiving thrombolysis suggests higher rates of recanalization with embolectomy, but lower rates of good functional outcomes. Given the lack of controlled studies to assess the impact of this treatment compared with alternatives, the use of embolectomy devices for acute stroke is considered experimental, investigational and unproven.