Article Text
PDF
Abstract
Objective Flexible large lumen aspiration catheters and stent retrievers have recently become available in the USA for the revascularization of large vessel occlusions presenting within the context of acute ischemic stroke (AIS). We describe a multicenter experience using a combined aspiration and stent retrieval technique for thrombectomy.
Design A retrospective analysis to identify patients receiving combined manual aspiration and stent retrieval for treatment of AIS between August 2012 and April 2013 at six high volume stroke centers was conducted. Outcome variables, including recanalization rate, post-treatment National Institutes of Health Stroke Scale (NIHSS) score, symptomatic intracranial hemorrhage, discharge 90 day modified Rankin Scale (mRS) score, and mortality were evaluated.
Results 105 patients were found that met the inclusion criteria for this retrospective study. Successful recanalization (Thrombolysis in Cerebral Infarction score 2B) was achieved in 92 (88%) of these patients. 44% of patients had favorable (mRS score 0–2) outcomes at 90 days. There were five (4.8%) symptomatic intracerebral hemorrhages and three procedure related deaths (2.9%).
Conclusions Mechanical thrombectomy utilizing combined manual aspiration with a stent retriever is an effective and safe strategy for endovascular recanalization of large vessel occlusions presenting within the context of AIS.
- Stroke
- Thrombectomy
- Stent
Statistics from Altmetric.com
Introduction
Recent comparison studies between endovascular techniques versus IV tissue plasminogen activator (tPA) for the treatment of acute ischemic stroke (AIS) have raised concern about the efficacy of endovascular therapy. However, these trials have not accounted for the rapid recent evolution of device technology and procedural technique.1–3 Stent retrievers (Solitaire FR, EV3 Neurovascular, Irvine, California, USA; Trevo, Stryker Neurovascular, Freemont, California, USA) have improved rates of recanalization, time to recanalization, and clinical outcomes compared with older generation thrombectomy devices (MERCI, Concentric Medical, Mountain View, California, USA).4 ,5 Aspiration thrombectomy, using flexible and highly navigable large bore distal access catheters, represents a second technique to efficiently achieve revascularization of large vessel occlusions.6 ,7
Endovascular and resultant clinical treatment failures may result from the inability to extract the occlusive thrombus, thrombus fragmentation with distal migration, and long time intervals to revascularization. We hypothesize that a combined technique using large bore aspiration catheters in conjunction with stent retrieval devices may improve recanalization rates. We present a retrospective multicenter experience using a combined stent retriever–aspiration thrombectomy technique.
Materials and methods
This multicenter retrospective study evaluated 105 consecutive patients treated for AIS between August 2012 and April 2013 at six institutions: Semmes-Murphey Clinic/University of Tennessee (Memphis, Tennessee, USA); the Medical University of South Carolina (Charleston, South Carolina, USA); Vanderbilt University Medical Center (Nashville, Tennessee, USA); Swedish Hospital (Denver, Colorado, USA); SUNY-Stonybrook (Stonybrook, New York, USA); and SUNY-Buffalo (Buffalo, New York, USA). Institutional review board approval was obtained at each participating institution. All patients undergoing endovascular treatment of AIS with a stent retrieval device and large bore aspiration catheter were included for evaluation. Consideration for AIS intervention was according to individual hospital practice. Radiographic evidence of large vessel occlusion was ascertained using screening CT angiography or digital subtraction angiography. Both anterior and posterior circulation strokes were included.
Procedure description
After angiographic visualization of an intraluminal thrombus, a guide catheter (Neuron or NeuronMax, Penumbra, Oakland California, USA; or Shuttle, Cook Medical Inc, Bloomington, Indiana, USA) was placed in the cervical arterial segment of interest. This was advanced as far distally in the vertebral or internal carotid artery as was safely feasible. In a triaxial fashion, a delivery microcatheter (Prowler Select Plus, Codman Neurovascular, Raynham, Massachusetts, USA; MERCI, Stryker; Rebar, EV3) was advanced through a large bore aspiration catheter (5MAX, 4MAX, 0.054, or 0.041 reperfusion catheter, Penumbra, Alameda, California, USA; Navien, EV3; DAC, Stryker Neurovascular) over a microguidewire. The largest size aspiration catheter that it was felt the vessel could accommodate was chosen. The microcatheter was advanced over the guidewire through the thrombus while the aspiration catheter was positioned just proximal to the thrombus. A stent retriever device (Solitaire FR or Trevo) was selected according to labeling indications and deployed through the microcatheter from distal to proximal across the site of occlusion. In most cases the delivery microcatheter was removed before thrombus extraction in order to allow for increased cross sectional luminal area for aspiration. Suction was either applied manually using a 20 mL or 60 mL syringe attached to the aspiration catheter via a Tuohy device or using the Penumbra aspiration pump apparatus. The stent retriever was then withdrawn into the large bore aspiration catheter during local aspiration or the stent retriever and aspiration catheter were slowly withdrawn together as a complete assembly. Subsequent angiographic runs were performed to evaluate revascularization, and the procedure was repeated as necessary. Further revascularization attempts, device utilization, lytic administration, and/or systemic anticoagulation were performed at the discretion of the treating physician.
Outcome measures
Outcome variables of interest included degree of vessel occlusion at presentation and post-treatment, defined by modified Thrombolysis in Cerebral Infarction (TICI) classification, symptomatic intracranial hemorrhage (SICH), discharge NIHSS, and functional outcomes obtained from neurology clinic records. Successful revascularization was defined as a TICI score ≥2B at the conclusion of the procedure. The post-treatment TICI score was determined by the treating physician at the time of the procedure. SICH was defined as presence of hemorrhage after treatment with worsening of clinical examination by ≥4 points on NIHSS. A good functional outcome was defined as a modified Rankin Scale (mRS) score of ≤2 at 90 days. Mortality was defined as death occurring within 90 days of initial presentation.
Data collection and analysis
Angiographic and clinical outcome data were collected and analyzed at each participating site. All protected health information removed data were submitted and compiled at the sponsoring institution (Semmes-Murphey Clinic). Statistical analysis was performed in Microsoft Excel.
Results
A total of 105 patients underwent embolectomy with the use of a stent retriever device and local aspiration (table 1). Mean age was 65 years and 60% (n=63) of patients were women. The median presenting NIHSS score was 17. Approximately 45% of patients (n=47) had received IV tPA. Mean time from onset of symptoms to groin puncture was 5.8 h. The sites of occlusion at the time of angiography included the right M1 (n=38), right M2 (n=3), right internal carotid artery terminus (n=9), left M1 (n=20), left M2 (n=5), left internal carotid artery terminus (n=20), and basilar artery (n=10). Two stent retriever devices were used: Solitaire (n=75) and Trevo (n=25). A combination of Solitaire and Trevo were used in five patients. The most commonly used distal access catheter was the Penumbra system (5Max, 4Max, or 0.054) followed by the Navien (n=4). The average number of passes required to achieve final recanalization was 2.3.
Numbers and proportions for selected patient characteristics
Thirty-seven per cent of patients achieved successful recanalization with one pass. The average time to recanalization from groin puncture was 57.3 min (range 11–240 min). The presenting TICI score was 0 in 101 patients, 1 in three patients, and 2a in one patient (outcome parameters are shown in table 2). A final TICI score of 2b or greater was achieved in 88% of patients. Forty-six patients (43.8%) had a TICI score of 3. Four patients failed to achieve acceptable recanalization (TICI 0–1).
Outcome characteristics
The median NIHSS score at discharge in surviving patients was 5. A total of 102 patients had a follow-up at 90 days. We chose to stratify patients based on mRS 0–2 and used it as our primary 90 day outcome measure as it is generally accepted as a well established marker of a ‘good outcome’. However, there are some patients for whom an mRS score of 3 would be considered an acceptable outcome rather than death. Thus a secondary analysis was also performed to stratify patients based on mRS ≤3. Among patients with follow-up, 45 (44.1%) had a favorable outcome (mRS 0–2); 28 patients (27.5%) had an mRS score of 3–5 (table 3).
Patient revascularization correlated to Thrombolysis in Cerebral Infarction score
Twenty-nine patients (28.4%) were dead (mRS=6) at 90 days; one secondary to pancreatic cancer and another to cardiac arrest. Three patients were lost to follow-up. There were five cases of SICH, three of which ended in death. One of these deaths occurred as the result of an arterial perforation during angioplasty of an underlying stenosis. In this case, the vessel was originally recanalized with the stent retriever and local aspiration but reoccluded due to an underlying stenosis. Two extracranial carotid dissections were reported. Six patients also received carotid stents, four of which were due to severe extracranial atherosclerotic disease. Five patients received angioplasty and two Wingspan (Stryker) stents were deployed. In 32 cases, IA thrombolytics or antiplatelets were used.
Discussion
Mechanical embolectomy devices have demonstrated increasing rates of recanalization and improved clinical outcomes with successive device iterations.4 ,5 ,8 Recent technological advancements have included the development of flexible large bore aspiration catheters and stent retrievers.
The Penumbra aspiration and distal access catheters (DAC, Stryker Neurovascular) were engineered to facilitate atraumatic navigation into the proximal intracranial vasculature. Studies of device efficacy for Penumbra aspiration showed a TIMI 2 or 3 recanalization rate of 81.6% with an 11% SICH rate. Favorable clinical outcomes (mRS 0–2) were seen in 25% of patients.9
Flexible intracranial stents have evolved over the past decade to non-detachable stent retriever devices, which can be delivered into the distal cerebrovasculature through microcatheters. This stent retriever configuration has improved on the ability of MERCI to engage and extract intracranial thrombus. After navigating beyond an occlusion with a microcatheter and wire, a stent retriever can be deployed across the entire length of the thrombus from distal to proximal. The inherent chronic outward force of the nitinol assembly results in circumferential displacement of the occlusive thrombus material, which subsequently becomes engaged within the interstices of the stent. The SWIFT trial, which evaluated the efficacy of the Solitaire stent retrieval device, showed a TIMI 2 or 3 recanalization rate of 61%. A favorable clinical outcome was seen in 58% of patients.5 The Trevo stent retrieval device showed a TICI 2 or 3 recanalization rate of 86% with an SICH rate of 7%. Favorable outcomes mRS (0–2) were seen in 40% of patients.4
All three of these thrombectomy systems (Penumbra, Solitaire, and Trevo) have shown improved technical efficacy, safety, and—in the case of the stent retriever devices—superior clinical outcomes when compared with MERCI. These improvements are likely a result of ease of use, speed of deployment, atraumatic design, and efficacy of clot removal.
The stent–aspiration technique combines the use of a large bore catheter with a stent retriever device, and this combination provides several potential synergistic effects. Localized aspiration at the site of the thrombus may promote entrapment of the thrombus within the stent. Flow control in the affected vascular territory may also reduce the incidence of thrombus fragmentation and distal embolization. The use of local aspiration may reduce or altogether eliminate emboli to new territories, which was observed in 7–9% of cases in the SWIFT and TREVO trials, and in up to 11% in subsequent registries.10 Removal of the delivery microcatheter from a 5max or 0.058 Navien catheter leaves a larger cross sectional surface area for the aspiration of thrombus, thereby substantially increasing the amount of suction which can be applied during the removal of the stent retriever device.
Withdrawal of the assembly is achieved by using the aspiration catheter as a conduit and pulling the stent retriever entirely into the aspiration catheter and subsequently out of the patient; an alternative technique is to simply apply local suction to the proximal edge of the thrombus with removal of the entire apparatus (aspiration catheter and stent retriever) as a unit with a portion of the distal stent and thrombus exposed. The first technique is possible with soft thrombus entrained entirely within the stent. This technique may also lessen the occurrence of thrombus fragmentation and is useful when removing the stent retriever through a proximal carotid stent. In addition, intermediate catheter access within the distal cerebrovasculature is preserved with this technique, obviating the need to renegotiate catheter position in the event that a second pass is needed, which is particularly helpful in cases with challenging cervicocerebral anatomy. It may not be possible to completely resheath a stent containing calcified or highly organized thrombus. In this scenario, the aspiration force applied to the catheter by the syringe is palpable to the operator and application of suction may help hold the thrombus in the stent. If there is a proximal carotid stent, the aspiration catheter also facilitates navigation of the guide catheter through the stent by smoothing the size differential between the guide catheter and microcatheter. The stent retriever and thrombus can then usually be housed within the larger guide catheter before removal to prevent carotid stent and stent retriever entanglement (figure 1).
(A) Lateral projection from a left common carotid angiogram demonstrates a high grade, flow limiting stenosis of the proximal cervical internal carotid artery (arrow) in a patient with an acute left middle cerebral artery (MCA) syndrome (National Institutes of Health Stroke Scale (NIHSS) score of 18) who failed to improve after IV tissue plasminogen activator administration. (B) Intracranial projections demonstrate an occlusive thrombus within the left MCA. (C) A cervical carotid stent was deployed over a wire using a distal protection device resulting in (D) proximal flow restoration. (E) A Penumbra 5max catheter was then positioned within the carotid siphon (arrow) and a MERCI microcatheter (18L) was navigated beyond the thrombus. (F) Superselective angiography confirmed the microcatheter position distal to the thrombus. (G) A Trevo device has been deployed spanning the thrombus (arrow). In addition to providing temporary bypass and reperfusion, the deployed Trevo stabilized the system within the MCA, facilitating navigation of the 5max into the MCA. Once the 5max had been navigated into the MCA, the 18L microcatheter was removed. The Trevo was then retracted with simultaneous aspiration of the 5max. The Trevo was completely retracted into the 5max aspiration catheter to prevent entanglement with the carotid stent during removal. (H) An elongated thrombus was retrieved from this thrombectomy pass. (I) Control angiography performed after thrombectomy (43 min after femoral puncture) demonstrated complete recanalization (Thrombolysis in Cerebral Infarction score of 3). The patient's NIHSS score was 0 at discharge.
Our results showed an overall TICI 2b/3 recanalization rate of 88%, SICH rate of 4.6%, average time to recanalization of 57.3 min, and average improvement in NIHSS score of 12. These rates compare favorably with those for stent alone.4 ,5 Extremely short revascularization times (<20 min) were possible, and the time to revascularization generally was dependent on proximal vascular access rather than ease of delivery of the stent retriever or aspiration catheter. The complications in this series occurred in two patients who had reperfusion-type intraparenchymal hemorrhages and one patient who had a vessel perforation related to angioplasty of an underlying atheromatous lesion. There were a total of six (5.7%) emboli to new territories using the stent aspiration technique. Two of these were emboli to distal arteries after successful recanalization of a proximal segment. The other four were emboli to a new territory (table 4). These rates compare favorably with previous reports using a balloon guide catheter and stent retriever alone.4 ,5
Location of initial occlusions and emboli to new territories
Weaknesses of this study are its retrospective nature and heterogeneity in patient selection. NIHSS and predetermined radiographic criteria, such as Alberta Stroke Program Early CT score and diffusion weighted imaging thresholds were not incorporated into the inclusion criteria for this study. Additionally, the method used to ascertain outcome measures relied heavily on retrospective data obtained from stroke registry databases and medical records, which in some cases is less than ideal and introduces the potential for bias. Further studies incorporating this technique should be prospective in nature, standardize inclusion criteria, and independently adjudicate angiographic results for recanalization and the presence of post-intervention distal territorial occlusions. Further prospective studies will also be needed to compare the outcomes and recanalization rates of this technique with patients undergoing more established thrombectomy techniques.
Lastly, because this was a multi-institutional study, there was some heterogeneity in the devices used for the employment of the stent–aspiration technique. The intent of this study was to focus on the technique rather than the specific type of device being used. The dominant technique used was the employment of a triaxial system utilizing a guide catheter and large bore aspiration catheter to deliver a stent retrieval device via microcatheter into an interlumial thrombus, thus allowing for both aspiration and stent assisted mechanical thrombectomy. The heterogeneity in the employment of the technique is an acknowledged limitation of the study in terms of generalizability. However, it can also be viewed as a potential benefit in that it provides a number of different options by which this technique can be successfully employed to achieve the desired result. Despite these limitations, this study does provide data that illustrate the safety and efficacy of the stent–aspiration technique and serves as a platform for future studies to evaluate the utility of this novel technique.
Conclusion
The findings of this study illustrate that technical nuances—in addition to rapidly evolving technology—impact angiographic and clinical outcomes. The combination of local aspiration and stent retrievers resulted in highly effective revascularization of large vessel occlusion in AIS patients.
References
Footnotes
-
Correction notice This article has been corrected since it was published Online First. David A Fiorella's name has been amended to read David Fiorella.
-
Acknowledgements The authors wish to thank Andrew J Gienapp for technical and copyediting help, preparation of the manuscript and figures for publishing, and publication assistance with this manuscript.
-
Contributors WH and VTD assisted with conception and design, acquisition of the data, analysis and interpretation of the data, drafting the article, critically revising the article, reviewed submitted version of the manuscript, and statistical analysis. DH, LE, and ASA assisted with conception and design, acquisition of the data, analysis and interpretation of the data, drafting the article, critically revising the article, reviewed submitted version of the manuscript, statistical analysis, and study supervision. DF, DL, GD, IC, JM, and RT assisted with acquisition of the data, analysis and interpretation of the data, drafting the article, and reviewed submitted version of the manuscript. AST, PM, AS, and MM assisted with acquisition of the data, analysis and interpretation of the data, drafting the article, critically revising the article, reviewed submitted version of the manuscript, and statistical analysis. DAF assisted with acquisition of the data, analysis and interpretation of the data, drafting the article, critically revising the article, reviewed submitted version of the manuscript, statistical analysis, and administrative/technical/material support.
-
Competing interests ASA reports other (consultancy) from Covidien, Johnson and Johnson, Siemens, Stryker, and Terumo; and grants from Siemens and Terumo, outside the submitted work. IC reports other (proctor) from Ev3/Covidien, outside the submitted work. DAF reports non-financial support (institutional support) from Siemens and Microvention; other (consultancy) from Coviden ev3, Cordis, NFocus Medical, and Micrus Endovascular; other (ownership interests) from Vascular Simulators LLC, TDC Technologies, and CVSL, outside the submitted work. DF reports other (modest stock options) from Penumbra Inc, outside the submitted work. MM reports grants from Toshiba, outside the submitted work. AST reports grants from Siemens, Microvention, Penumbra, Covidien, and Stryker, outside the submitted work. RT reports grants from Siemens, Microvention, Penumbra, Covidien, and Stryker, outside the submitted work.
-
Ethics approval The study was approved by the institutional review boards of the University of Tennessee/Semmes-Murphey, Medical University of South Carolina, Vanderbilt University Medical Center, Swedish Hospital, SUNY-Stonybrook, and SUNY-Buffalo.
-
Provenance and peer review Not commissioned; externally peer reviewed.