Objective To elucidate the early detachment of the Solitaire stent during mechanical thrombectomy, we tested the stent in vitro under various conditions and evaluated with microscopy.
Methods A total of five stents were tested. Using a dynamic mechanical analysis machine, we calculated the tensile force needed for separation of the four stents from the pusher wire. Stent Nos 1 and 2 were tested without any prior manipulation. During the retraction test, no sheathing around the stent–wire junction was applied to stent No 1 whereas stent No 2 was partially covered with a sheath in accordance with instructions for use (IFU) recommendations. Stent No 3 was tested after deployment and retrieval were performed three times in an in vitro vascular replica with sheathing during retrieval. Stent No 4 was tested after one deployment and retrieval using the same replica as in stent No 3 without sheathing. In addition, forward pushing force was applied intentionally during retraction to induce excessive bending of the stent. Stent No 5 was pulled apart by both hands. After separation, stent tips and pusher wires were studied for all stents under the microscope.
Results The tensile force needed for separation was about 6 N for stent Nos 1, 2, and 3, and about 4 N for stent No 4. All of the stents showed separation at the proximal marker, not at the detachment zone.
Conclusions Detachment of the Solitaire stent during thrombectomy can be due to separation around or inside the proximal marker. Adherence to the manufacturer's IFU of partial re-sheathing during retraction and not using the device for more than two passes might decrease the possibility of such device failure.
Statistics from Altmetric.com
Mechanical thrombectomy using stent retriever has shown better angiographic outcomes than previous devices.1 It also has the merit of restoring flow as soon as it is deployed. Among the stent retrievers available, the Solitaire AB/FR device is the first fully retrievable and electrolytically detachable self-expanding intracranial stent.2 ,3 Mechanical thrombectomy using this device has shown relatively better results with less vessel injury compared with previous devices.4–8
Early detachment or separation of the Solitaire FR device has been reported in previous studies (table 1).5–7 ,9 ,10 There have been 39 similar events registered at the Manufacturer and User Facility Device Experience (MAUDE) on the US Food and Drug Administration website in less than 1 year after launching of the Solitaire FR in the USA in 2012.11 Most of the reported cases have shown poor outcomes and without successful retrieval of the accidentally detached devices. Considering the fact that this device was originally made to be detached for aneurysm embolization assistance, it can be hypothesized that this device can be detached inadvertently during retraction, especially when an entangled thrombus and stenotic vessel lumen create more resistance. To date, there has been no specific elucidation or investigation on the risk and mechanism of early detachment or separation. Here we conducted an in vitro investigation of the stent under various conditions using a dynamic mechanical analysis (DMA) machine and microscope.
A DMA (DMA Q800; TA Instruments, New Castle, Delaware, USA), which has a force resolution of 0.00001 N and a strain resolution of 1 nm, was used to explore the force required to detach the Solitaire stent from a pusher wire. All mechanical examinations were conducted using a tensile clamp at room temperature. During each test, the stent part was fixed to the clamp and the pusher wire was retracted by machine while recording the amount of tensile force continuously (figure 1A). Stent Nos 1 and 2 were tested without any prior manipulation. During the DMA tests, no sheathing was applied to stent No 1 (Solitaire AB 6×20), while partial sheathing around the proximal marker band (from the distal wire to the proximal stent struts) was applied to stent No 2 (Solitaire AB 6×20), as recommended by the instructions for use (IFU). Stent No 3 (Solitaire AB 6×30) was deployed and retrieved three times in an in vitro vascular replica, which mimicked the human internal carotid artery/middle cerebral artery, before the DMA test.12 We deployed stent No 3 in the distal middle cerebral artery portion and retracted it through the siphon to the proximal internal carotid artery. Stent No 3 was partially re-sheathed, like stent No 2, during retraction. Stent No 4 (Solitaire AB 6×30) was deployed and retracted once with no partial re-sheathing, as employed with stent No 1. Further, one round of forward pushing force was applied at the siphon segment to intentionally cause excessive bending around the proximal marker of stent No 4 (figure 1B). We did not use any thrombus-like material to retrieve with stent Nos 3 and 4. For stent No 5 (Solitaire AB 4×20), we did not use the DMA machine; rather it was pulled apart by manual force without re-sheathing by one of the authors. We compared the results of this manual manipulation with the DMA induced stent separation results to determine the differences (table 2).
The stress–strain curves for stent Nos 1, 2, 3, and 4 are presented in figure 2. The force required to break all of the specimens was <6 N, and the tensile force needed to detach stent No 4 was about two thirds (about 4 N) of the force observed for the other stents (stent Nos 1, 2, and 3) (table 2).
Microscopic views of the separated stents are shown in figure 3. None of the tested stents showed detachment at the detachment zone. Stent Nos 1, 2, and 3 were separated at the proximal adhesive. The ball and socket joint was dislocated, and there was a ball tip at the end of the pusher wire. Stent No 4 was separated at the distal adhesive, and the ball and socket joint were dislocated. The proximal marker was attached at the end of the pusher wire, and the proximal adhesive also seemed to have failed. Stent No 5 was separated due to fracture at the junction point of the ball and socket joint (table 2).
The Solitaire AB/FR is the first fully retrievable and detachable intracranial stent made by laser cutting from a nitinol hypotube. This closed cell design stent has an open slit along the long axis which enables the stent to be folded in the vessels.2 Its advantage of being able to fold on itself inside smaller intracranial vessels and its complete retrievability make it an effective tool for mechanical thrombectomy cases. The stent tapers in its proximal portion where it is connected to the pusher wire. Mechanically, the stent and pusher wire are connected inside the proximal marker through a ball and socket joint. The proximal end of the stent comprises the socket and the tip of the pusher wire ends as a ball. The proximal marker band wraps around the joint, and each end of the band is fixed to the proximal stent struts distally and the pusher wire proximally by adhesive.13 The detachment zone, where electrolysis occurs, is located proximal to the proximal marker band and adhesive (figure 3B).
Despite the positive clinical results of the Solitaire thrombectomy device, inadvertent mechanical disconnections of the stent from the pusher wire during thrombectomy procedures have been reported.5–7 ,9 ,10 Although most of the cases showed a poor clinical outcome, no specific description or investigation regarding the mechanical aspect of the device has been reported. According to the MAUDE website, the stent involved in the event was not returned to the company for evaluation as it was implanted in the patient and is presumed to be early detachment at the detachment zone in most of the 39 cases reported.11 However, considering that none of the stents in this investigation showed separation at the detachment zone, some previously reported early detachments could have been failure of the adhesive and dislocation of the ball and socket joint inside the proximal marker. The proximal marker is fixed to the stent strut distally and the pusher wire proximally using an adhesive instead of welding.13 Therefore, separation at either side of the proximal marker may result if the adhesive's resistance to the tensile force is lower than the resistance to the tensile force at the detachment zone. In the case of early proximal adhesive disruption by the tensile strength applied, the proximal marker will be found to be attached at the stent and the pusher wire with the ball tip, as in stent Nos 1, 2, and 3 in this experiment. In the case of distal adhesive failure, we might see the proximal marker attached at the end of the pusher wire, as in stent No 4. The ball and socket joint can be either simply dislocated as in stent Nos 1, 2, and 3, or even fractured at the junction point of the ball and pusher wire. as in stent No 5. Without examination under the microscope, it is almost impossible for the operator to know the existence of ball tip at the end of the remaining pusher wire because of their small size. That is why we believe that all of the MAUDE registration cases were reported as simple early detachment.
According to the IFU from the maker, it is recommended that the operator should re-sheath the proximal part of the stent with the microcatheter during retrieval, as in stent Nos 2 and 3 in this study.14 In this experiment, there was no significant tensile strength difference between stent Nos 1, 2, and 3. From a comparison of the results for stent Nos 1 and 2, we can presume that partial re-sheathing at the straight vessel segment does not greatly influence the tensile strength of the Solitaire stent. Considering that the results of stent No 3 were similar to stent Nos 1 and 2, three trials of deployment and retrieval within the vessel replica with partial re-sheathing also does not significantly reduce the tensile strength of the stent. However, the result was different when the specimen was retrieved without re-sheathing through the curved vessel replica and even significantly bended by pushing the device forward, as in stent No 4. We believe that partial re-sheathing with the microcatheter prevents the stent–wire junction from over bending and protects the junction around the proximal marker from failure, especially at the curved or tortuous segment during retrieval, as in stent No 3. However, significant bending around the proximal marker band, as in stent No 4, can induce failure of the adhesive and weaken the device junction, making it prone to separation. Hence prevention of the over bending during retraction by re-sheathing the microcatheter sufficiently over the proximal marker and only pulling backwards during retrieval, as per the IFU, can reduce the chance of mechanical failure of the adhesive and prevent the chance of inadvertent separation of the stent.6 This may be especially true when retrieving the clot laden device through curved and/or stenotic segments. Also, as per the manufacturer's suggestions, not torquing the device with the stent deployed in the thrombus is also important for a similar reason.
The company also recommends that the integrity of the stent should be checked regularly after every pass and not to use one device for more than two flow restoration attempts.14 Although stent No 3 did not show a significant difference in force at break (N) relative to stent Nos 1 and 2 (table 2, figure 2), four of the 39 MAUDE cases were reported to be separated during the third attempts.11 Therefore, we believe that strict adhesion to the IFU is important in reducing the occurrence of early detachment. Findings of calcification or severe atherosclerotic stenosis at the thrombosed or proximal segment may also predispose to consequences such as stent entrapment.10 Also, recapture of the device with the original or larger catheter should always be considered when more than the usual resistance is perceived during retraction.
During the DMA retraction test, straight tensile force was applied in vitro at room temperature, which is different from real clinical applications, and can affect the function of the stent's adhesive. Therefore, the results will not be valid or relevant if the difference is significant. Also, due to limited funds, we carried out this investigation with a small number of stents, restricting our ability to make definitive conclusions; however, this study supports the concept that the Solitaire stent can separate around or inside the proximal marker.
Detachment of the Solitaire stent during thrombectomy can be due to separation around or inside the proximal marker. Adhesion to the manufacturer’s IFU of partial re-sheathing to avoid excessive bending and not using the device for more than two passes might decrease the possibility of such device failure.
The authors thank Noriaki Matsubara and Thomas F Flood for their help in the manuscript preparation.
Contributors H-JK: conception, design, experiment, draft, and revision. J-YC: experiment, draft, and revision. ASP: draft and revision. H-SK: conception, revision, and final approval of the version to be published.
Funding This work was supported by the Research Fund of Chungnam National University, grant No 2012-1670.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.
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.