Safety and efficacy of mechanical thrombectomy with the Solitaire device in large artery occlusion
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.100701
Source of Support: None, Conflict of Interest: None
Background and Purpose: Intravenous tissue plasminogen activator (TPA) has limited efficacy in proximal large vessel occlusions. This study was to assess the safety and efficacy of mechanical thrombectomy with a retrievable Solitaire stent in acute large artery occlusions . Materials and Methods: This is a single center study enrolling patients treated with Solitaire-assisted thrombectomy between November 2010 and March 2011. Inclusion criteria were severe stroke of National Institutes of Health Stroke Scale (NIHSS) score ≥10, treatment initiation within 6 hours from onset, and an angiographically verified occlusion of proximal middle cerebral artery (MCA) or internal carotid artery (ICA). The primary outcome was recanalization defined as Thrombolysis in Cerebral Infarct (TICI) reperfusion grade 2b/3. Secondary outcomes were good functional outcome at 3 months (modified Rankin Scale [mRS] ≤2), early substantial neurological improvement (NIHSS score improvement ≥8 at 24 hours), and symptomatic hemorrhagic transformation (SHT). Results: Ten patients were consecutively enrolled: Age: 72.4 5.7 years; female: 70%; baseline median NIHSS score: 19.5; and ICA occlusion in 50% and M1 portion of MCA occlusion in 50%. Six patients received intravenous TPA before intra-arterial treatment, and five patients were treated with adjuvant intra-arterial urokinase. Successful recanalization was achieved in 7 (70%) patients. Four (40%) patients had a good functional outcome at 3 months, and three (30%) patients had an early substantial neurological improvement. SHT occurred in two patients (20%), and 3-month mortality rate was 30%. There was no procedure-related complication. Conclusions: Mechanical thrombectomy with the Solitaire device can effectively recanalize proximal large vessel occlusions, and potentially improves clinical outcome.
Keywords: Stent, stroke, thrombectomy, thrombolysis
Acute stroke due to a large vessel occlusion is a catastrophe, usually destined to death or severe disability. With intravenous tissue plasminogen activator (tPA) therapy, the recanalization rate was only 10% in internal carotid artery (ICA) occlusion and less than 30% for proximal middle cerebral artery (M1 portion of MCA) occlusion. , For severe strokes of NIHSS score ≥16, even after intravenous TPA, one third died, another one third remained severely disabled (modified Rankin Scale [mRS] 4-5), and only one-fourth of patients were independent (mRS 0-2) at 3 months. If not treated with TPA, the chance of good outcome further decreased to 18%. ..
Intra-arterial (IA) thrombolysis is a promising option for acute large arterial occlusions. Several trials and a meta-analysis have indicated that IA fibrinolysis substantially increases recanalization rates and good and excellent clinical outcomes. ,, However, a large thrombus is quite resistant to a standard microcatheter thrombolysis technique and usually require a prolonged time for recanalization, which would fail to improve outcomes even after achieving recanalization. ,,, Accordingly, new thrombectomy devices have been developed to recanalize large vessel occlusions more successfully and more rapidly. ,,,, The Solitaire AB (ev3 Inc, Plymouth, MN) is a self-expanding and fully retrievable stent that can be used as a thrombectomy device for large arterial occlusions. Recent studies have shown encouraging results of high recanalization rates. , This study is to report safety and efficacy of the Solitaire thrombectomy device in patients with severe acute ischemic stroke due to proximal large vessel occlusions.
This is a single center study of mechanical thrombectomy with the Solitaire device (Solitaire AB; ev3 Inc, Plymouth, MN), and patients were enrolled between November 2010 and March 2011. Our thrombolysis protocol for patient selection is given in [Figure 1]. Inclusion criteria were (1) age ≥18 years, (2) baseline NIHSS ≥10, (3) angiographically confirmed ICA or M1 occlusion, (4) symptom duration ≤6 hours, and (5) no contraindication for reperfusion therapy. Intravenous TPA was given to patients when symptom duration was less than 4.5 hours based on the European Cooperative Acute Stroke Study (ECASS) III trial results.  However, intravenous TPA dose was modified to 0.6 mg/kg according to the protocol of Interventional Management of Stroke (IMS) trial. , Informed consent was obtained from patients, family members, or surrogates.
Using digital subtraction angiography, we identified target vessels. We placed a 7F long sheath (Shuttle, Cook) or 8F-9F guide catheter (Envoy, Cordis) on the cervical segment of ICA or distal common carotid artery. The clot was carefully passed with a 0.014-inch guide wire (Traxcess, ev3 or Agility, Cordis). A microcatheter (Rebar-18 ev3, or Prowler Select Plus, Cordis) was navigated over the wire just distal to the thrombus under fluoroscopic control. After placement of microcatheter beyond the thrombus, the 4 × 20 mm Solitaire device was advanced through the microcatheter and deployed to cover the entire vessel segment where the thrombus lodged. Then, we performed an angiogram to confirm successful stent unfolding and to assess immediate flow re-establishment. The stent device was kept in place for 1 to 5 minutes and then slowly retrieved. We could not use a balloon catheter to achieve flow arrest specially designed for the Solitaire device because it was not formally available in Korea. Instead, to prevent re-embolization, manual negative suction with a 50 cc syringe and simultaneous carotid compression were applied during stent retrieval. If the occlusion persisted after the first pass, the procedure was repeated and adjuvant IA urokinase infusion was considered. Maximum attempts of stent deployment and use of IA urokinase were left at the discretion of responsible neurologist and interventionalist. IA therapy was terminated when intracerebral hemorrhage was suspected or 7 hours passed from symptom onset. During the procedure, we collected data of time intervals of groin puncture, initiation and end of thrombectomy, and recanalization from symptom onset, urokinase dose, and numbers of the Solitaire device passes.
Patient evaluation and outcome measure
NIHSS scores were assessed on admission, at 24 hours after procedure and discharge, and whenever patients had a neurological deterioration. Functional outcome was assessed with mRS at discharge and at 90 days. Pre-stroke mRS disability was also obtained from interviewing family members. Computed tomography (CT) or magnetic resonance imaging (MRI) scan was performed in all patients before and 24 hours after reperfusion therapy. When patients had a neurological deterioration, an emergent CT scan was obtained.
The primary outcome was recanalization defined as Thrombolysis In Cerebral Infarction (TICI) reperfusion grade 2b or 3 on immediate post-procedural angiograms.  Secondary outcomes were good functional outcome at 3 months (mRS ≤2) and early substantial neurological improvement (NIHSS score improvement ≥8 at 24 hours). Safety outcomes were (1) symptomatic hemorrhagic transformation (SHT) defined as intracranial hemorrhage on CT/MRI associated with NIHSS score decline ≥4 within 24 hours, (2) device-related complications of vessel perforation, arterial dissection, or embolization to an initially uninvolved territory, and (3) any hemorrhagic transformation on follow-up imaging classified by the ECASS criteria.
Ten consecutive patients (mean age: 72.4 ± 5.7 years, 7 (70%) females) were treated. Median baseline NIHSS score was 19.5 (range, 13 to 27) and there were 5 M1 and 5 ICA occlusions. Six patients (60%) received intravenous tPA before IA therapy, and five (50%) patients were treated with adjuvant IA urokinase (dose range, 40,000-200,000 U) to further recanalize distal branches. Duration of IA procedure was 80.5 ± 24.0 minutes, and the interval from onset to treatment completion was 248.0 ± 76.4 minutes. Summary of patient characteristics and the outcome of each patient are presented in [Table 1] and [Table 2].
Successful recanalization was achieved in 7 (70%) patients (3 M1 and 4 ICA occlusions): The TICI reperfusion grade was 3 in four patients and 2b in three patients. Median number of the Solitaire passing was 1.5 times (range, 1-3). Recanalization was achieved after one pass in five, 2 passes in one, and 3 passes in one patient.
At 90 days, four (40%) patients had a good functional outcome (mRS 0-2). Three-month functional outcome was associated with recanalization. Of the seven recanalized patients, 4 (57.1%) had a good mRS outcome, whereas 3 patients without recanalization had a poor outcome (mRS 4 in one patient and mRS 6 in two patients) [Figure 2].
Three (30%) patients had an early substantial neurological improvement of NIHSS score decrease ≥8 at 24 hours. These three patients had a TICI reperfusion grade 3 and an excellent outcome at 90 days (mRS 0 in one and mRS 1 in two patients). Mortality rate at 3-month was 30%. The causes of death were herniation, aspiration pneumonia, and undetermined, respectively. Clinical efficacy and safety outcome is summarized in [Table 3].
Any hemorrhagic transformation within 24 hours occurred in six patients (2 hemorrhagic infarction type 1, one hemorrhagic infarction type 2, and 3 parenchymal hematoma type 2). SHT developed in two patients. One case was a 77 year-old woman with ICA occlusion and initial NIHSS score of 22. She failed to achieve recanalization, and neither intravenous TPA nor IA urokinase was administered. She died because of herniation at 3 days from stroke onset. The other case was a 74-year-old man with M1 occlusion and initial NIHSS score of 19. He received intravenous TPA and had a recanalization after thrombectomy without adjuvant IA urokinase. This patient had a decompressive surgery, but he died after discharge to a long-term care facility. In addition to two SHTs, one parenchymal hematoma type 2 was detected on follow-up CT, but it was not associated with a neurological deterioration. None of the case had a device-related complication.
This study shows that the Solitaire device-assisted thrombectomy is a promising option for large vessel occlusions. Successful recanalization was achieved in 70%, which led to a good outcome in 40% and an early substantial neurological improvement in 30% of all treated patients. Single pass was sufficient to recanalize the primary occlusion site in five patients. Adjuvant IA urokinase was required in five patients, but these patients had recanalization from a single pass and IA urokinase was delivered to achieve a better reperfusion in distal vascular beds. Since thrombectomy alone substantially reduced thrombus burden, the required urokinase dose was small as 4 of 5 patients received ≤100,000 U (mean dose: 50,000 U). None of IA urokinase-treated patients developed SHT.
Our recanalization rate is higher than Mechanical Embolus Removal in Cerebral Ischemia (MERCI) trial result, similar to those of Prolyse in Acute Cerebral Thromboembolism (PROACT) II, IMS II, and Multi-MERCI trials, and less than those of Penumbra trial and another Solitaire study [Table 4]. ,,,, The frequency of good outcome in the current study is greater than those of MERCI and Penumbra trials and comparable to those of PROACT II, IMS II, Multi-MERCI trials, and another Solitaire study. It is of note that our patients were older (≥70 years in 80%) than those enrolled in other IA reperfusion studies. Considering that elderly patients are known to have poor response to treatments, our outcome result is quite encouraging.
Recently, results from the SWIFT (Solitaire With the Intention For Thrombectomy) trial comparing the efficacy and safety of Solitaire and Merci devices were presented at 2012 International Stroke Conference  The Solitaire group had a higher recanalization rate and achieved more good outcomes of mRS 0-2 than did the Merci group. The rates of successful recanalization (88.9%) and good outcome (58.2%) in the Solitaire group of the SWIFT trial were higher than our results. The discrepancies between the SWIFT trial and our study might reflect the difference between clinical trial and real world practice. Since clinical trial results are usually derived from highly experienced experts and selected patients, real world practice frequently fails to achieve a treatment efficacy as much as seen in clinical trials. In addition, although our patients were treated earlier than those in the SWIFT trial (167.5 vs 294 minutes), the higher age (average 72.4) and higher NIHSS score (median 19.5) of our patients as compared to those enrolled in the SWIFT trial (average age, 67; NIHSS 17.3) partially accounts for the outcome discrepancy.
To compare with intravenous tPA results, we extracted data of comparable patients from the National Institute of Neurological Disorders and Stroke (NINDS)-tPA trial database. Since baseline NIHSS scores of our patients were ≥16 except for one patient (NIHSS score of 13), we selected TPA-treated patients who had a NIHSS score of 16 or more. In the comparable NINDS trial's tPA-treated patients, mRS 0-2 at 3 months was observed in only 26%, which is much lower than our finding. When further restricted to patients with age ≥70 and NIHSS ≥16, the rate of mRS 0-2 further decreased to 16%. ,, A prior conventional angiography study demonstrated that more than 90% of patients with NIHSS ≥12 had an occlusion in ICA, M1, or basilar artery.  Therefore, the low rates of good outcome with intravenous TPA in patients with severe stroke must be attributed to a low recanalization rate in large vessel occlusions. , These data suggest that for severe stroke with large vessel occlusion IA fibrinolysis should be considered.
The SHT of 20% in this study is higher than in other studies, but the mortality rate of 30% is comparable [Table 4]. Of the two SHTs, one SHT occurred in a patient who received neither intravenous TPA nor intra-arterial thrombolytic agent. He failed to achieve recanalization with thrombectomy and had a malignant hemispheric infarction involving whole ICA territory with massive hemorrhagic conversion. Therefore, the SHT in this patient might be attributed to a treatment failure rather than a thrombectomy-related complication. For asymptomatic hemorrhagic transformation, our frequency of 40% was modestly higher than those of other IA therapy studies and much greater than 6% of comparable NINDS cohorts.
In conclusion, the Solitaire device-assisted thrombectomy can achieve a high recanalization rate and potentially improves functional outcome with an acceptable safety in patients with large vessel occlusions. Well-designed trials should be undertaken to add this promising treatment to the arsenal of proven therapies for acute stroke.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4]