Article Text
PDF
Abstract
Background To place a stent retriever for thrombectomy in acute ischemic stroke, the clot has to be passed first. A microwire is usually used for this maneuver. As an alternative, a wireless microcatheter can be used to pass the clot.
Objective To analyze the feasibility and complication rates of passing the clot using either a microwire or a wireless microcatheter.
Methods A retrospective non-randomized analysis of 110 consecutive patients with acute ischemic stroke in the anterior circulation was performed, in whom video recordings of mechanical thrombectomies were available. In total, 203 attempts at mechanical recanalization were performed.
Results Successful recanalization (TICI 2b–3) was achieved in 97.3% of patients. In 71.8% of attempts the clot was successfully passed using a wireless microcatheter only. When a microwire was used initially, clot passage was successful in 95.3% of attempts. Complication rates for angiographically detectable subarachnoid hemorrhage were 6.1% when a microwire was used to pass the clot compared with 0% when a wireless microcatheter was used (p<0.001). Complication rates for angiographically occult circumscribed subarachnoid contrast extravasation observed on post-interventional CT scans were 18.2% when a microwire was used to pass the clot and 4.5% when a wireless microcatheter was used (p<0.001).
Conclusions In most cases of mechanical recanalization the clot can be passed with a wireless microcatheter instead of a microwire. In our study this method significantly reduced the risk for vessel perforation and subarachnoid hemorrhage. We therefore recommend the use of this technique whenever possible.
- thrombectomy
- stroke
Statistics from Altmetric.com
Introduction
A variety of techniques of endovascular mechanical recanalization have been shown to be effective in acute ischemic stroke.1–4 The most common technique is mechanical thrombectomy with stent retrievers. To place the stent retriever it is necessary to pass the clot with a microcatheter first. The standard technique for this maneuver is to position the microcatheter proximal to the clot, then pass the clot with a microwire, and finally advance the microcatheter over the microwire. Passing the occlusion site and/or using a stent retriever is associated with a risk of perforating the occluded vessel leading to contrast extravasation and subarachnoid hemorrhage.5–9 To the best of our knowledge, there are no studies in which the effects of passing the clot versus extracting the clot have been independently evaluated.
As an alternative to passing the occlusion site with a microwire, a wireless microcatheter can be used for this maneuver. Specifically, the microcatheter is advanced over a microwire almost to the proximal face of the clot. The microwire is then retracted into the microcatheter until the tip of the microwire lies about 10 mm proximal to the tip of the microcatheter. In this way, the microwire still stabilizes the microcatheter but the tip of the microcatheter is soft and can flex when it meets resistance. In this configuration the microcatheter is advanced until it has passed the clot.
In this paper we analyzed the feasibility and complication rates of passing the clot in patients with acute ischemic stroke using either a microwire or a wireless microcatheter.
Materials and methods
Study set-up and patient cohort
This single-center retrospective analysis was conducted on endovascular stroke treatments performed between November 2016 and November 2017. At the start of the study period we had begun to obtain video recordings of fluoroscopic and X-ray images of the interventional procedures in accordance with internal guidelines of our hospital and department. All fluoroscopy and X-ray images shown on the monitor of the angiographic system (Siemens Artis zee, Erlangen, Germany) were recorded using video documentation (ICUE Player, Teracue, Odelzhausen-Munich, Germany). For this study, all video recordings and X-ray images were archived on our PACS system. All videos, images, and reports of the procedures were used for analysis. Ethics approval for the retrospective analyses was granted by the ethics board of our institution.
In the study period a total of 173 consecutive patients were admitted to our department for mechanical thrombectomy in the anterior circulation. We excluded 13 patients in whom the intervention was stopped without thrombectomy since the occlusion site had already been recanalized after intravenous thrombolysis. Fifty patients underwent mechanical recanalization without video recording since the intervention was performed in angiography suites outside the Department of Neuroradiology. This left 110 patients to be included in our study.
Stroke management and endovascular techniques
Details on stroke management at our institution and the treatment regimens which the patients receive have been published previously.7 Data on materials used and techniques performed are shown in tables 1–4.
Details of attempts at clot passage with regard to different anatomical locations of vessel occlusions
Types of microcatheters used in clot passes and success rates of passing the clot using the wireless microcatheter technique
Types of microwires used during all thrombectomy attempts (n=203), for primary or secondary microwire-first clot passage (n=99), and in microwire-first cases with angiographically occult contrast extravasations (SCE, n=18) or angiographically confirmed vessel perforations (n=6)
Vessel perforations with angiographically observed contrast extravasation during mechanical thrombectomy
The decision as to whether the initial attempt to pass the clot was made using a microwire or a wireless microcatheter was at the interventional neuroradiologists' discretion. Types of microcatheters and microwires were selected according to the preference of the interventionalist, depending on the location of the clot and the anatomical configuration of the occluded vessel. When first-pass microcatheter passage was not successful, the microwire was used secondarily to pass the clot. It is standard practice in our institution to perform a control angiography series when the technique is changed, and to perform a microcatheter injection after successful clot passage. Therefore we can reliably determine the timing and location of potential complications.
All patients received a post-interventional CT examination performed on a 16-slice CT scanner (Siemens SOMATOM Definition AS, Straton MX P, Erlangen, Germany) immediately after the end of the thrombectomy procedure prior to transportation to the intensive care unit.
Data analysis and statistics
Patient demographics are given as mean±SD or median and range depending on data distribution.
We assessed the localization of vessel occlusion, attempts at mechanical recanalization performed, and all materials used for intervention. All imaging data of the mechanical recanalization (video data of the endovascular intervention, saved fluoroscopy and X-ray images) were analyzed with a special focus on the passage of the clot and on peri-interventional complications, especially subarachnoid contrast extravasations.
Post-interventional CT scans were analyzed in randomized order on standard PACS workstations by two experienced neuroradiologists who were blinded to all clinical data. Areas of hyperdensities were evaluated and classified into the following categories: parenchymal hemorrhage (coagulum), leakage of the blood-brain barrier, or subarachnoid contrast extravasation.
Primary endpoints were defined as successful clot passage. Secondary endpoints were defined as subarachnoid contrast extravasations, observed either during angiography and confirmed on post-interventional CT scans (angiographically detectable subarachnoid hemorrhage) or angiographically occult circumscribed subarachnoid contrast extravasation observed on post-interventional CT scans.
Data on clot passage and peri/post-interventional subarachnoid contrast extravasation as well as materials used were analyzed using χ2 tests, Fisher exact tests, and Student’s t-tests. P values ≤0.05 were considered statistically significant. All analyses were performed using SPSS 22.0 (IBM SPSS, Chicago, Illinois, USA).
Results
Patients
One hundred and ten patients (60 women) with an acute ischemic stroke in the anterior circulation were analyzed. In total, 203 mechanical recanalization attempts were performed. The mean age of the patients was 73.5±14.2 years (range 18–94). No significant differences in distribution of age and gender were found in the patients.
Location of clots
The anatomical distribution of the vessel occlusions is shown in table 1. Seventy-eight patients had a single clot. Five patients had simultaneous vessel occlusions with clots in multiple independent vascular territories—for example, in the middle and anterior cerebral arteries—requiring several attempts at recanalization. In 27 patients multiple clots were present in consecutive vessel segments, which therefore required multiple recanalization attempts. In 11 of these 27 patients the proximal clot was located in the distal carotid artery with additional clots in the anterior and middle cerebral arteries. In the remaining 16 patients the proximal clot was located in the M1 segment of the middle cerebral artery with additional independent emboli in consecutive M2 segments. For methodological reasons it was not possible to determine if all clots in consecutive vessel segments were already present at the start of the intervention or if some of these clots were the result of peri-interventional embolism.
Recanalization rates and number of recanalization attempts
A total of 140 distinct occluded vessel segments in 110 patients were treated by endovascular recanalization.
Successful recanalization as defined by a Thrombolysis in Cerebral Infarction (TICI) score of 2b–3 was achieved in 97.3% (107/110) of patients.8 9 Full recanalization (TICI 3) was achieved in 61.8% of patients (68/110), 35.5% had TICI 2b (39/110), two patients had TICI 2a, and one patient had TICI 1.
When analyzed per patient (n=110), an average of 1.8 recanalization attempts were performed (range 1–5). In 56.4% of patients recanalization was successful after a single pass (62/110), and in 20% after a second pass (22/110). Fourteen patients needed three attempts, 11 patients four attempts, and one patient had five recanalization attempts.
When analyzed per occluded vessel segment (n=140), an average of 1.45 recanalization attempts were performed. The first-pass success rate amounted to 67.9% (95/140) and the second-pass success rate to 27.9% (30/140).
Passage of clot
In 124 of the 203 mechanical recanalization attempts the interventionalist initially tried to pass the occlusion site with a wireless microcatheter (wireless microcatheter technique). In 71.8% of attempts (89/124) the clot was successfully passed using a wireless microcatheter only.
In 28.2% of attempts (35/124), passing the occlusion site was not possible with a wireless microcatheter only. The maneuver was therefore modified and a microwire was used secondarily. Passing the clot with a microwire was then successful in 100% of attempts (35/35).
The success rate of passing the clot with a wireless microcatheter at first pass showed no significant differences with regard to the anatomical location of the clot (p=0.98, table 1). Detailed information on first-pass rates of different microcatheters used are given in table 2.
In 67 of the 203 mechanical recanalization attempts the microwire was initially used to pass the occlusion site (microwire-first technique). Passage was successful in 95.5% of attempts (64/67).
In three cases of peripheral clots (anterior cerebral artery A2 segment, middle cerebral artery M2/3 segment), passing the clot was not possible and intervention was stopped.
In 12 of the 203 attempts an intermittent technique was performed combining microcatheter and microwire (ie, the clot was partially passed with a wireless microcatheter and partially using a microwire). These 12 attempts were excluded from further data analysis.
Complications
In six patients angiographically visible contrast extravasation corresponding to active subarachnoid hemorrhage was diagnosed during the intervention (6/110; 5.5%). With reference to the number of attempted clot passes, the complication rate amounted to 3.0% (6/203). In none of the attempts when a wireless microcatheter was used to pass the occlusion site did active bleeding occur (0%). All extravasations diagnosed during the intervention occurred when a microwire was used to pass the clot (6/6; 100%). This difference was statistically significant (p<0.001). Thus, the complication rates for severe subarachnoid hemorrhage were 6.1% (6/99) when a microwire was used initially or secondarily to pass the clot versus 0% (0/89) when a wireless microcatheter was successfully used (p<0.001). In four of these six cases the microwire was used primarily. In the remaining two cases the microwire was used secondarily because the initial attempt to pass the clot with the microcatheter was not successful. In both cases when the microwire was used secondarily, the clot was located at the carotid-T while the perforation occurred in an M2 segment after secondary microwire probation. Therefore we conclude that the perforation cannot be related to the unsuccessful prior attempt to pass the clot with the microcatheter. Thus the complication rates for severe subarachnoid hemorrhage were 6.0% (4/67) when a microwire was used initially or 5.7% (2/35) when a microwire was used secondarily to pass the clot.
In four of these six patients bleeding stopped spontaneously and no further treatment was required. In one of the remaining two patients there was active hemorrhage in the M3 segment of the middle cerebral artery and endovascular coil embolization was performed (figure 1). In the other patient contrast extravasation was seen only following a microcatheter injection in a M3 segment distal to the clot. Since the clot was still present and effected an occlusion of the middle cerebral artery, this was not considered to be active bleeding requiring treatment. The interventionalist therefore decided to stop the procedure and leave the remaining clot in the M2 segment. However, since the vessel was definitely perforated, we regarded this as a subarachnoid contrast extravasation. The locations of the perforations are shown in table 4.
Vessel perforation with angiographically observed contrast extravasation in an elderly patient admitted 50 min after onset of acute ischemic stroke with dysarthria and left-sided hemiparesis. On admission the M1 segment of the right middle cerebral artery was occluded. After intravenous thrombolysis, two clots in a frontal and parietal M2 segmental artery remained (upper left). The clot in the frontal M2 segment was passed effortlessly at first attempt using a Synchro soft wire followed by a Trevo 18 microcatheter. After removal of the microwire, a microcatheter injection was performed with evidence of contrast extravasation indicating vessel perforation (upper right). Coil embolization was required to stop active bleeding. Afterwards, mechanical thrombectomy of the parietal M2 branch was performed successfully using the Solumbra technique (TICI 2b) (lower left). On post-interventional CT scans, contrast extravasation was observed in the right subarachnoid cisterns (lower right).
Angiographically occult ipsilateral circumscribed subarachnoid contrast extravasation was identified on post-interventional CT scans in 22 additional patients (22/110; 20%). A typical example is shown in figure 2. In none of these cases were clinical symptoms related to the subarachnoid contrast extravasation observed. In 18 of these 22 patients the clot was passed using a microwire and in four of the 22 patients using a wireless microcatheter. This difference was statistically significant (p<0.001). The complication rate for angiographically occult circumscribed subarachnoid contrast extravasation was significantly higher (p<0.001) with 18.2% (18/99) when a microwire was used to pass the clot compared with 4.5% (4/89) when a wireless microcatheter was used.
Angiographically occult contrast extravasation observed on post-interventional CT scans in an elderly patient admitted 2 hours after onset of acute ischemic stroke with right-sided hemiparesis and global aphasia. On admission the proximal M1 segment of the left middle cerebral artery was occluded (left). After intravenous thrombolysis, two clots in the distal M1 segment and a peripheral M2 segment remained. Successful mechanical recanalization with five attempts using the SAVE and the Solumbra technique resulted in full recanalization (TICI 3) (middle). No evidence of contrast extravasation was observed on angiographic images. On post-interventional CT scans, discrete contrast extravasation was observed in the left Sylvian fissure (right).
In total, subarachnoid contrast extravasations were observed in either angiographic or CT images in 28 patients. In 24 of these patients the clot was passed with a microwire and in four patients with a wireless microcatheter. Complication rates related to microwire passage were 24.2% (24/99), with 18.8% on initial microwire passage (12/64) and 34.3% on secondary microwire passage (12/35), compared with a significantly lower complication rate of 4.5% in wireless microcatheter passage (4/89) (p<0.001).
One case of parenchymal bleeding was diagnosed on post-interventional CT in the right temporal lobe (1/110; 0.9%). This was considered unrelated to the endovascular procedure.
Leakage of the blood-brain barrier in the infarcted brain areas was seen in 64.5% of the patients (71/110). There was no significant correlation between subarachnoid contrast extravasation, parenchymal bleeding, and leakage of the blood-brain barrier.
Effect of materials
Detailed information on microcatheters used for mechanical thrombectomy, successful passage rates and associated complications is given in table 2. The type of microcatheter had no significant impact on successful clot passage (p=0.830).
Detailed information on microwires used for mechanical thrombectomy and associated complications is given in table 3. Most complications (5/6 perforations observed during angiography and 20/24 circumscribed subarachnoid contrast extravasations observed on CT) occurred when a microwire with a standard (not soft or atraumatic) tip was used. After correction for the number of attempts, the overall complication rates were 25.6% (20/78) for microwires with a standard tip versus 19.0% (4/21) for a microwire with a soft tip. However, this difference was not statistically significant (p=0.20).
Discussion
Endovascular thrombectomy has been shown to be an effective and safe treatment for ischemic stroke. However, several procedure-related complications are associated with this therapy. The reported rates of procedure-related complications vary between 4% and 31%.10 Vessel perforation and dissection are considered to be the most severe complications. Countless variations of the thrombectomy technique have been described, but the procedural step of passing the clot has not been addressed in detail to date.
Our results suggest that it is significantly safer to pass the clot with a wireless microcatheter than with a microwire, and that passing the clot with a wireless microcatheter is feasible in more than 70% of passages.
That a wireless microcatheter is safer may seem counterintuitive at first sight as the tip of the microcatheter is larger than the tip of the microwire. The explanation is probably as follows. From observations in animal studies (unpublished data), the authors conclude that microwires or microcatheters are not being pushed through an intravascular clot but rather pass between the clot and the vessel wall. The microwire or microcatheter is advanced with similar force. When the tool meets resistance—for example, because the vessel bifurcates and the tool runs against the vessel wall—this force works at the vessel wall. The strain exerted on the vessel wall then depends not only on the absolute magnitude of force but also on the area of the vessel wall on which it works. A 0.014 inch microwire, as used by many interventionalists, has an outer diameter of 0.3556 mm and covers an area of 0.10 mm2. A typical microcatheter used in many thrombectomies (eg, Rebar 18) has an outer diameter of 0.81 mm and covers an area of 0.52 mm2. The force at the tip of either the microwire or microcatheter is the product of applied pressure and cross-sectional area of the respective device. Consequently, when pushed against the vessel wall with the same pressure, the microwire results in a fivefold larger force on the affected portion of the vessel wall than the microcatheter and may thus be more traumatic.
An alternative explanation is that vessels vulnerable to mechanical damage like the lenticulostriate arteries originating from the M1 segment of the middle cerebral artery or small perforators originating from the M2 or M3 segments have diameters that are usually smaller than the diameter of microcatheters. This means that microwires may enter such vessels unnoticed when pushed through a clot whereas microcatheters will not.
These considerations are evidenced by our observation that the complication rates for major and minor subarachnoid contrast extravasation were 6.1% and 18.2% when a microwire was used to pass the clot, but only 0% and 4.5% when a wireless microcatheter was used.
Our results make it plausible that different sizes and types of microcatheters or microwires may also have an effect on complication rates. In our study the use of microwires with standard tips resulted in a higher complication rate than the use of microwires with soft tips (15.8% vs 11.1%). This difference was not statistically significant, most likely because our sample size was too small, and should therefore be reconfirmed in a larger study. However, it is a potential benefit of our study that it draws attention to the fact that microcatheters and wires so far have not been specifically optimized with regard to crossing ability and safety in thrombectomy. It is conceivable that optimized microcatheters will achieve even higher success rates for clot passage than those we found.
In our study we used two different definitions for peri-interventional vessel injury. Angiographically observed contrast extravasations after passage of the clot can be clearly attributed to the maneuver of passing the clot. In this series our rate of 5.5% for this complication was higher than the average reported in the literature, or the rate reported previously by our own group.5 6 11 This may be explained by the peripheral locations of the vessel occlusions in this study collective. Five of the six perforations were located in M2, M3, or A3 segments of the cerebral arteries, which are associated with a higher risk of periprocedural complications. Only one perforation occurred when a clot in the M1 segment was passed.
As a second measure for peri-interventional vessel injury, we used angiographically occult circumscribed subarachnoid contrast extravasation observed in post-interventional CT. These either result from iatrogenic vessel perforation or from microtrauma in terms of endoluminal trauma or shear forces on the perforator vessels when a stent retriever is used.12 Microtrauma during clot extraction cannot be prevented since the stent retriever is an essential part of the mechanical recanalization maneuver, and is always associated with shear forces to the vessel. Our finding of 20% of patients with angiographically occult subarachnoid contrast extravasations is consistent with previously published data.8 9 It has also been reported that the risk for this complication increases with the number of thrombectomy attempts, and many interventionalists tend to ascribe this complication to the effects of stent retriever use.8 9
With respect to this, the results of our study are surprising. Using a wireless microcatheter to pass the clot reduced the risk for angiographically occult subarachnoid contrast extravasations from 18.2% to 4.5%. Hence our data suggest that, in most cases, these contrast extravasations are not caused by stent retrievers but by the use of a microwire to pass the clot. The previously reported increased risk after multiple thrombectomy maneuvers thus appears not to be a consequence of repeated stent retriever use, but rather of repeated maneuvers to pass the clot.
If microwires were used in this study, they usually had tip shapes between 45° and 60°. In some centers J-shaped microwires are routinely used to cross occluded vessel segments because they are believed to be less traumatic. Indeed, J-shaped microwires should be less likely to enter perforating side branches while passing the clot. However, there is also a potential disadvantage. From observations in animal studies (unpublished data), the authors are aware that when a J-shaped microwire is advanced into distal vessel segments with smaller diameters (distal to the clot), the radius of the J decreases. This leads to increased rigidity of the curved tip, increased strain on the vessel wall, and potential dissection. To the best of our knowledge, so far there are no published data to conclude which tip shape should be used.
Study limitations
This study suffers from several shortcomings: (1) Most importantly, this is a retrospective analysis and the number of patients with ischemic stroke undergoing endovascular mechanical recanalization with video surveillance is limited. However, it reports the largest patient group undergoing cerebral thrombectomy in the anterior circulation with video and CT surveillance to date. (2) In this non-randomized study, the choice between microcatheter and microwire to pass the clot was at the discretion of the interventionalist. However, this study was performed at a high-volume thrombectomy center. All participating interventionalists had long-standing experience in a variety of recanalization techniques and were extensively trained in both techniques to pass the clot. In any case, our study results can be used to justify and design randomized follow-up studies. (3) Our sample size is too small to allow for definitive conclusions regarding different types of microcatheters or microwires.
Conclusions
Our data suggest that, in most cases of mechanical recanalization, the clot can be passed with a wireless microcatheter instead of a microwire. In our study this method significantly reduced the risk for vessel perforation and subarachnoid hemorrhage. We therefore recommed the use of this technique whenever possible.
Our study draws attention to the fact that microcatheters and wires have not so far been specifically optimized with regard to crossing ability and safety in thrombectomy.
Footnotes
Contributors AK designed the study, collected and analysed the data, drafted and revised the manuscript, approved the final version, and agrees to be accountable for all aspects of the work. ON designed the study, drafted and revised the manuscript, approved the final version, and agrees to be accountable for all aspects of the work. AM made substantial contributions to the interpretation of data for the work, revised the draft paper, approved the final version, and agrees to be accountable for all aspects of the work. WDS made substantial contributions to the interpretation of data for the work, revised the draft paper, approved the final version, and agrees to be accountable for all aspects of the work. MW conceptualized the work, designed the study, analysed the data, drafted and revised the manuscript, approved the final version, and agrees to be accountable for all aspects of the work. He is the guarantor.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests MW reports personal fees from Stryker Neurovascular, Siemens Healthcare, Bracco Imaging, and Medtronic outside the submitted work and has received grants for educational exhibits (non-personal) from the following companies: Abbott, ab medica, Acandis, Bayer, Bracco Imaging, B Braun, Codman Neurovascular, Kaneka Pharmaceuticals, Medtronic, Dahlhausen, Microvention, Penumbra, Phenox, Philips Healthcare, Route 92, Siemens Healthcare, SilkRoad Medical, St Jude, Stryker Neurovascular. AM reports personal fees from Neuravi, Penumbra, Stryker, and Perflow outside the submitted work.
Patient consent Not required.
Ethics approval Ethics Committee, University Hospital Aachen RWTH.
Provenance and peer review Not commissioned; externally peer reviewed.
Correction notice Since this paper was first published online the author Wilson D Scott has been corrected to Scott D Wilson.