Article Text
PDF
Abstract
Background Formation of clot fragments during mechanical thrombectomy for acute ischemic stroke can occlude the distal vasculature, which may reduce the rate of good clinical outcome.
Objective To examine the hypothesis that distal embolization can be reduced using stent retriever thrombectomy in combination with Lazarus Cover technology.
Methods Hard, fragment-prone clots were used to create middle cerebral artery occlusions in a vascular phantom. Three different treatment strategies using Solitaire FR included: group 1—proximal flow control with an 8F balloon guide catheter (BGC), group 2—thrombectomy through a 6F conventional guide catheter (CGC), and group 3—a similar thrombectomy procedure to group 2 but including the Lazarus Cover device. The primary endpoint was distal emboli quantified by the number and size of the clot debris.
Results The Cover-assisted stent retriever thrombectomy significantly reduced the generation of clot fragments >200 μm as compared with thrombectomy with a CGC, and was similar to the BGC group. Particle size distribution <200 μm was similar across the groups. All groups were associated with high rates of recanalization, with only one failed recanalization with partial clot retention after three passes in one experiment of stent retriever thrombectomy through a CGC. Use of the adjunctive Cover device did not prolong the procedure as compared with control groups.
Conclusions For a fragment-prone clot, Solitaire thrombectomy in conjunction with the Cover device may lower the risk of distal embolization and is comparable to BGC-protected embolectomy.
- Thrombectomy
- Stroke
- Device
Statistics from Altmetric.com
Introduction
New evidence from recent randomized control trials has shown a convincing benefit from stent retriever-mediated mechanical thrombectomy for selected patients with acute ischemic stroke (AIS) with large vessel occlusion in the anterior circulation.1–4 The SWIFT PRIME results showed that in the intervention group, for every 2.6 patients who were treated, one additional patient had an improved disability outcome; for every four patients who were treated, one additional patient was functionally independent at the 90-day follow-up.4 These findings suggest that a new standard of care for AIS management has been established.5
Although the latest generation of clot retrieval systems has been shown to be an effective AIS treatment, the rate of successful recanalization (59–88%) in endovascular trials has not always mirrored the rate of good clinical outcomes (33–60%).1–4 The use of clot retrievers can cause clot fragmentation with the release of distal emboli.6 ,7 The presence of fragmented clots is a predictor of poor outcome8 as disrupted clots can block collateral flow to potentially salvageable tissue or cause distal embolization in a previously unaffected area.9–12 Procedural release of embolic particulate is a modifiable risk.
Recently, a new adjunctive technology for mechanical thrombectomy has been introduced that serves as an invertible mesh to encapsulate the clot as the stent retriever is extracted, thereby reducing the risk of distal embolization as the clot is mobilized out of the vasculature.13 ,14 It is our hypothesis that the safety of thrombectomy devices as indicated by the risk of embolic shower may potentially be altered by the use of the Cover device. The objective of this study was the characterization of distal emboli generated during Solitaire thrombectomy in combination with the Lazarus Cover (Medtronic Neurovascular, Irvine, California, USA).
Materials and methods
Middle cerebral artery occlusion model
The model system was composed of a human vascular replica, clot model, and physiologically relevant mock circulation flow loop. The vascular replica was built using a small-batch manufacturing process described elsewhere.15 Patient-specific 3D reconstruction of the vasculature from CT angiography was modified to include only the internal carotid artery (ICA), posterior communicating artery, the anterior cerebral artery (ACA), and middle cerebral artery (MCA) including the M2 divisions. The MCA had an average diameter of 3 mm. The two M2 divisions were designed to rejoin distally, resulting in a single output. A core-shell mold of the aforesaid model was created and 3D printed using fused deposition modeling. The core-shell mold was immersed in a heated aqueous solution of sodium hydroxide for removal of the support material. A degassed mixture of Sylgard 184 silicone elastomer (Dow Corning, Midland, Michigan, USA) was infused into the core-shell mold and cured at 70°C overnight. By dissolving the core-shell mold in xylene after silicone curing, a transparent and flexible silicone replica was obtained. The inner wall of the resulting replica was lubricated by coating dual layers of LSR topcoat (Momentive Performance Materials, Albany, New York, USA) to reduce the friction between the device and the vascular phantom.
The hard, inelastic clot was generated by thrombin-induced clotting of bovine blood (2.5 NIHU thrombin/mL blood) with addition of barium sulfate (1 g/10 mL blood). This specific clot model was chosen because it is inelastic and prone to fragment,16 which allowed us to simulate the mechanical thrombectomy procedure in the worst case scenario. A segment of 4 mm×5 mm (diameter×length) clot was injected into the flow loop into the silicone replica using a 35 mL Kendall Monoject catheter tip syringe to form an MCA occlusion. The width of the tip at base is ¼ inch (6.4 mm), which is bigger than the diameter of the clot (4 mm) to prevent clot fragmentation.
The silicone replica was connected into the flow loop with continuous monitoring of MCA flow and pressure to ensure physiologically accurate hemodynamics, as previously described.17 The flow loop contained a peristaltic pump, which was used to deliver an oscillatory flow of saline solution, and hosecock compressor clamps to provide peripheral resistance. By adjusting the peripheral resistance, the baseline flow conditions were matched to normal human phase contrast MR measurements for each of the vessels, creating a hemodynamically representative cerebrovascular model. Saline which traveled through the external carotid artery was directed to a filtration system before re-entering the reservoir; on the other hand, saline flowing through the MCA and ACA was collected in a container for particle analysis.
Flow restoration procedure
A 4 mm×20 mm Solitaire FR device (Medtronic Neurovascular) was used to achieve recanalization according to the manufacturer's instructions. Three treatment strategies were tested including: (1) Solitaire thrombectomy through an 8Fr Cello balloon guide catheter (BGC; Medtronic Neurovascular) with the tip placed at the cervical ICA (control No 1), (2) Solitaire thrombectomy through a 6Fr Envoy conventional guide catheter (CGC; Cordis, Miami Lakes, Florida, USA) with the tip placed at the cervical ICA (control No 2), and (3) Solitaire thrombectomy with the Lazarus Cover device deployed through a 5Fr Navien (Medtronic Neurovascular). In both control groups, aspiration during thrombectomy was performed with two 25 cc syringes connected to a three-way stopcock through either the BGC or CGC. Aspiration was not performed in the Cover group. In the Cover group, a triaxial system was used including a 6Fr Shuttle Sheath proximally (Cook Corporation, Bloomington, Indiana, USA) and a Navien positioned at the origin of the MCA. After deployment of the Solitaire, the microcatheter was removed and the rapid-exchange Cover was delivered to the proximal MCA. The Navien was withdrawn to unsheathe the Cover device. After device deployment a stopwatch was used to ensure that in all groups 4 min elapsed before initiating thrombectomy.
Eight experiments were carried out for each treatment strategy. Experiments were done by a block randomization. Specifically, each block consisted of one experiment from each group and the order of the experiments in the block was randomly assigned. The maximum number of thrombectomy attempts was limited to three.
Particulate analysis for distal emboli
Clot fragments generated during thrombectomy and collected into two reservoirs, one for emboli to the MCA distribution and the other to the ACA distribution, were analyzed. After thrombectomy, clot fragments (>1000 µm) were first separated from the collected sample by a 10 mL graduated pipette. The number and the Feret diameter of these fragments were recorded. Characterization of the emboli smaller than 1000 µm was conducted using the Colter principle (Multisizer 4 Coulter Counter; Beckman Coulter, Brea, California, USA). All particulate analysis was performed by an operator blinded to the experimental group.
Two aperture diameters used in this study were 400 and 2000 μm, and each aperture can be used to measure particles within a size range of 2–60% of its diameter. After primary catheter deployment, 700 mL saline was collected as blank for particle characterization, and the results were subtracted from the following tests. The emboli generated in this study covered a wider range than a single aperture can measure; therefore, two apertures were used to provide a complete particle size distribution analysis (8–1000 μm). Analysis of distal emboli was split into five categories: fragments with a diameter (1) >1 mm, (2) 200–999.9 µm, (3) 100–199.9 µm, (4) 50–99.9 µm, and (5) 8–49.9 µm. The number of the clot fragments with size >200 µm was summed to yield the size distribution owing to the infrequency of the observation, whereas repeat measurements from a single experiment for the smaller particles are reported as average.
Endpoints
The primary endpoint was the number and size of distal clot fragments generated during the procedure. A secondary endpoint was the recanalization rate and time to full flow restoration.
Statistical analysis
Results were expressed as mean±SD. Data were analyzed using the Prism (GraphPad Software). A check was first made for normality of distribution. If normality was confirmed, then one-way analysis of variance was used. If the distributions were not normal, then a non-parametric Kruskal–Wallis test was used. Significance was concluded when p<0.05.
Results
Number of emboli with size >1000 µm
The risk of forming clot fragments >1000 µm was significantly increased with the use of the CGC (p<0.05, online supplementary material video I) as compared with the BGC and the Cover (online supplementary material video II). An example of Solitaire thrombectomy with the use of CGC is shown in figure 1. Formation of large emboli with size >1000 µm occurred in seven out of eight cases in the CGC group, compared with two or three out of eight experiments in the Cover and BGC groups, respectively. There was no significant difference in the number of large emboli (>1000 µm) between the BGC and the Cover (p>0.05). Figure 2A shows that four out of five (80%) large emboli found in the Cover group were generated from one experiment (experiment 4). The size distribution of all large emboli regardless of treatment group ranged from 1 to 5 mm (figure 2B). It was observed that during mechanical thrombectomy, the metal-mesh Cover device slid over the Solitaire retriever and wrapped the retriever with the clot fragments inside (figure 3).
Representative control case of Solitaire thrombectomy via a conventional guide catheter with fragmentation immediately following initiation of the first pass (A, arrowhead). (B and C) Clot migration and fragmentation occur during clot removal. (D) One clot fragment is retrieved along with the Solitaire FR (arrow), and the other clot fragment is carried to the anterior cerebral artery by the antegrade flow (arrowhead). Arrows highlight clot engaged within the stent retriever and arrowheads point to fragments leading to distal embolization.
(A) Comparisons of emboli number (>1000 µm) between three different clot removal strategies. Most emboli are found in the conventional guide catheter (CGC) group (*p<0.05). (B) The size of these emboli ranges from 1 to 5 mm. BGC, balloon guide catheter.
(A) In a representative Cover experiment the clot is engaged by the Solitaire FR after deployment. Inset: opened configuration of the Cover device before Solitaire retrieval. (B–D) Clot debris is retained within the Cover and the risk of distal embolization is reduced. The inset in (C) shows closed configuration of the Cover device during Solitaire retrieval. Clot fragments captured by the Cover device are presented in the inset in (D). The arrows indicate clot location.
Supplementary video 1
Stentriever thrombectomy with a conventional guide catheter Example of large fragments of radiopaque clot becoming dislodged during the thrombectomy procedure using only a conventional guide catheter (CGC, 6Fr Envoy) in the cervical internal carotid artery. As the video begins, initial engagement shows large clot that was not mobilized by the device. Clot fragments are seen to be carried with the flow distally to the middle cerebral artery distribution during withdrawal of the stentriever through the tortuous carotid siphon and as the device is pulled into the guide catheter.
Supplementary video 2
Cover-assisted stentriever thrombectomy The stentriever is deployed and thrombectomy begins. As the device-clot system is pulled into the siphon, the clot is seen to slowly migrate toward the distal end of the stentriever. However, as the Cover device inverts and encapsulates the clot; the clot becomes fixed to the stent surface leading to complete thrombectomy with no visible distal emboli.
Number of emboli with size between 200 and <1000 µm
Similarly, in the size range between 200 and <1000 µm, thrombectomy with the use of CGC generated the most distal emboli compared with all the other treatment strategies (p<0.05) (figure 4A). The Cover technique had the smallest number of emboli with size between 200 and 1000 µm; however, no significant difference was found between the Cover and the BGC approaches. The average size of the emboli within this size range was 237.4 µm.
(A) Total number of emboli with size between 200 and <1000 µm collected during the experiments (*p<0.05). Average number of emboli with sizes 8–49.9 µm, 50–99.9 µm, and 100–199.9 µm generated during the procedure is presented in (B), (C), and (D), respectively. BGC, balloon guide catheter; CGC, conventional guide catheter.
Number of emboli with size <200 µm
On average, in the size range <50 µm, mechanical thrombectomy via BGC showed a trend towards reduction of microscopic clot fragments, followed by the Cover and CGC (p>0.05) (figure 4B). Emboli within this size range accounted for >90% of total emboli generated during the thrombectomy procedure regardless of the technique employed (BGC: 7644/7671, CGC: 15 418/15 460, Cover: 12 420/12 445). The Cover system, as compared with the BGC, tended to be more efficient in reducing the risk of 50–99.9 µm (figure 4C) and 100–199.9 µm (figure 4D) clot fragmentation (p>0.05 in both size ranges).
Distribution of clot fragments
Approximately half of the disrupted clots with size >1 mm were found in the previously unaffected area (ACA, 55%) in the CGC group, compared with 40% in the Cover group and 33% in the BGC group. On average, 80% of the clot fragments with size between 200 and <1000 µm travelled through the MCA. In the size range between 8 and 200 µm, the clot fragments were evenly distributed into the MCA and ACA regardless of the treatment approaches (47% in the ACA vs 53% in the MCA).
Recanalization rate and flow restoration
A complete restoration of flow was seen in all cases except one in the CGC group. In one out of eight experiments with the use of CGC, a clot fragment was retained in one of the MCA branches after three attempts at mechanical thrombectomy. Successful clot removal was achieved at the first attempt in all other cases. Time from device deployment across the clot to complete recanalization was <6 min regardless of the technique applied (5 min 39 s with the CGC, 5 min 41 s with the Cover, and 5 min 49 s with the BGC).
The pre-occlusion MCA and ACA flow rate was 132.7±2.5 and 61.8±1.5 mL/min, respectively. Complete occlusion of the MCA was confirmed by both flow reduction acquired by the flow sensor and angiography following clot injection. For the immediate flow restoration after Solitaire deployment, there was a trend for the CGC and Cover to offer increases in flow (p>0.05). The average MCA flow after device deployment was 45.0±12.2, 52.4±9.7, and 51.9±16.3 mL/min for the BGC, CGC, and Cover groups, respectively.
When the ICA flow was blocked by the BGC, applying aspiration resulted in flow reversal (−25.6±40.6 mL/min). Aspiration through the CGC did not cause flow reversal (53±15.1 mL/min). In the Cover group, no aspiration was applied, and the MCA flow during clot retrieval was 86.1±14.1 mL/min.
Discussion
Mechanical thrombectomy in an in vitro model system of cerebrovascular occlusion provides new insight into therapeutic strategies for efficient removal of clot burdens. To date, several adjunctive endovascular techniques are used in acute stroke interventions to reduce the risk of distal embolization such as BGCs, which have been shown to be an independent predictor of favorable clinical outcomes.18 In this study, Solitaire thrombectomy was performed in conjunction with the use of an embolic protection device, the Cover device. Outcome measures including characterization of distal emboli and quantification of flow restoration during the procedure were recorded.
Rapid reperfusion of ischemic penumbra is associated with favorable outcomes and reduced mortality.19 ,20 For every 30 min delay in reperfusion, there is a 10% relative reduction in the probability of good clinical outcomes.21 Good clinical outcomes, defined as a modified Rankin Scale score of 0–2, were also found to be strongly associated with complete reperfusion.22 ,23 A Thrombolysis in Cerebral Infarction (TICI) score of 3 is the desired result of endovascular AIS treatments, and is assessed when complete perfusion is achieved, without clot fragmentation and distal embolization. Although the success of endovascular revascularization is often assessed by a variety of angiographic grading systems,24 ,25 small microemboli may not be seen on imaging, and their effect on the clinical outcomes may be underestimated.
Our results demonstrate that the new Cover assisted thrombectomy is comparable to the BGC technique in its risk of distal embolization, time to recanalization, number of thrombectomy attempts, and recanalization rate despite using a smaller access approach and not significantly altering afferent flow. Importantly, 80% of distal emboli >1 mm were found in only one experiment in the Cover group, suggesting the relatively low occurrence of the most damaging clot fragments with this adjunctive device. Moreover, in our study we did not apply aspiration through the guide catheter during Cover-enhanced thrombectomy in an effort to quantify the characteristics of the device alone. The embolic protection of the Cover device described herein could potentially be increased when coupled with aspiration. The Cover technique is statistically better than the CGC method in several hard clot subgroups. Our study uses robust determination of particle size distribution in the effluent and is consistent with preliminary reports that the Cover device prevents distal emboli events.13 ,14 ,26
The Cover device is designed to minimize loss or fragmentation of clot during stent retriever thrombectomy, and early clinical data in 10 patients (six M1, three M2, and one tandem occlusions) have been reported.26 TICI 2b–3 recanalization was obtained after only one pass in six out of nine M1 or M2 occlusion cases. None of the 10 patients had embolization of the previously unaffected area. On the contrary, the presence of emboli in the previously uninvolved territories (ACA) was observed in previous in vitro studies13 ,14 and in our study. Mokin et al13 compared the effectiveness of the Cover device for stent retriever thrombectomy with the BGC and the CGC techniques in an in vitro MCA model. It was shown that stent retriever thrombectomy with the Cover device achieved a higher rate of successful recanalization (91%), defined as TICI 2b–3, as compared with stent retriever thrombectomy with the BGC (45%) or the CGC (50%) (p<0.05). No significant differences were noticed between the BGC and the CGC group in the rate of achieving TICI 2b–3. Overall, the recanalization rate reported in Mokin's study was lower than that reported in our study. This discrepancy might be related to the different clot types and the different materials of the vascular replicas used in the two studies.
Over the course of the thrombectomy procedure, clot fragments may become dislodged and lead to downstream emboli owing to manipulation of an endovascular device such as a guidewire, microcatheter, or stent retriever. Current clot characterization protocol can provide only cumulative emboli data collected over the entire duration of the procedure. A reproducible in vitro set-up that can monitor emboli formation continuously during mechanical thrombectomy is needed to acquire instantaneous time-resolved data showing when the clot fragments are formed.
To date, several adjunctive endovascular techniques are used off-label during acute stroke interventions. We assessed the effectiveness of A Direct Aspiration first Pass (ADAPT) technique for stroke thrombectomy recently.27 In that study, we chose not to use aspiration during Cover-enhanced thrombectomy to understand purely the contribution of the device towards emboli protection. However, we expect that clinical practice will entail aspiration during Cover-enhanced thrombectomy that will further serve to reduce distal emboli.
Our study has some limitations. During the thrombectomy procedure, the ACA flow was into the reservoir for capture of distal emboli. Therefore, there was no collateral path in our system representative of the anterior communicating artery, which might have augmented the performance of BGC-mediated aspiration and also altered the distribution of distal emboli. In addition, it is known that the performance of thrombectomy is, in part, affected by the clot mechanics, and we tested this technology on only one clot model. Finally, our model overestimated the recanalization rate of stent retriever thrombectomy (96%) as compared with recent clinical trial results.
Summary
The risk of embolic shower can potentially be altered by the use of adjunctive devices that serve to encapsulate the occlusive clot. For a hard, inelastic clot, the use of the Cover technique may reduce the risk of distal embolization and is comparable to adjunctive BGC. Future work will seek to determine if Cover-assisted stent retriever thrombectomy with aspiration via a distal access catheter under proximal flow control provides a benefit by reducing distal emboli. More clinical data are expected to confirm the safety and efficacy of the Cover device.
References
Footnotes
Contributors J-YC: designed and performed the experiments, analyzed and processed the data, drafted the manuscript. ASP and MJG: designed the study, performed the experiments, revised the draft manuscript.
Funding This study was supported by Medtronic Neurovascular. The content is solely the responsibility of the authors, and does not necessarily represent the official views of Medtronic Neurovascular.
Competing interests MJG: has been a consultant on a fee per hour basis for Codman Neurovascular and Stryker Neurovascular; holds stock in InNeuroCo; and has received research support from the National Institutes of Health (NIH), Codman Neurovascular, Stryker Neurovascular, Microvention, Medtronic Neurovascular, Philips Healthcare, InNeuroCo, Neuronal Protection Systems, the Wyss Institute and Silk Road. ASP: holds stock in InNeuroCo, and has received research grants from Medtronic Neurovascular and Stryker Neurovascular.
Provenance and peer review Not commissioned; externally peer reviewed.