A Novel Thrombectomy Device: An In Vitro Evaluation of a Prototype Catheter

Purpose: This prototype catheter is a newly-developed distal access catheter featuring a self-expanding, flexible, funnel-shaped tip. The purpose of its design is to reduce the risk of thrombus fragmentation during mechanical thrombectomy and improve first-pass recanalization (TICI 3). In this experimental setup, we preclinically evaluated the effectiveness and navigability of the new catheter. Methods: A vessel model was filled with a blood-like-viscous medium, and the image was projected with the corresponding vessel area by camera transmission to correspond to the conditions in an angiography. Thrombi from porcine blood were placed into the Arteria Carotis interna of the vascular model and subsequently mechanically thrombectomized with a stent retriever. In the first part, the prototype was compared to a standard distal-access-catheter without using an external catheter. (N = 20 for each catheter). In the second part the prototype was inserted through a guiding catheter (n=11) to determine the navigability performance. Results: In the first experimental series, mechanical thrombectomy was successful 19 out of 20 times (95% success rate) for the prototype catheter versus 15 out of 20 times (75% success rate) for the standard distal-access-catheter. In the second experimental series, the prototype catheter achieved first-pass recanalization 10 out of 11 times (91 % success rate) and 1 out of 11 times at second pass (9%). Conclusion: This series of experiments demonstrated higher first-pass recanalization rates for the newly-developed funnel-shaped prototype featuring a self-expanding tip in comparison to a cylindrical standard distal-access-catheter.


Introduction
Mechanical thrombectomy (MT) has become the standard treatment for acute ischemic stroke (AIS) in large vessel occlusions (LVO) of the anterior circulation [1] . Several studies have demonstrated the benefits using MT in combination with IV rt-PA lysis.
However, improved long-term functional outcome, defined as modified Ranking Scale (mRS) < 2, was registered in only 33-60% of cases. This could be caused by clot fragmentation during MT resulting in embolization of peripheral vessels in up to 17% of cases and embolization in new territories (ENT) in 2 to 11% of cases.
Clot fragmentation during MT may occur at different times, during stent retriever retraction before reaching the catheter tip or at the wall of the catheter. To the best of our knowledge, this problem still remains unresolved. Parts of the thrombus can be carried to the periphery during MT, thereby occluding collateral vessels. This can often cause irreversible damage to the brain parenchyma. This is associated with worsening neurological outcome and increased mortality rates [7][8][9] . The rate of TICI 3 recanalizations can be considered as a suitable and clinically relevant safety measure for studies evaluating MT devices and techniques [10] .
Thrombus loss or fragmentation of the thrombus at the catheter wall could be caused by the cylindrical shape of a standard distal-access-catheter (DAC) which does not expand across the entire vessel lumen. The disproportionally larger size of the selfexpanding stent retriever in relation to the smaller opening of a standard DAC during retraction may lead to clot abruption at the wall of the tip. Currently, only cylindrically shaped catheters are available, which cannot be adjusted flexibly to the inner diameter of the distal end. Therefore, the Department of Neuroradiology at the University Hospital Mainz, Germany, developed a novel catheter designed to prevent thrombus loss and fragmentation at the catheter wall during stent retriever retraction. A recent proof-ofprinciple-study showed a better recanalization rate for the funnel-shaped tip compared to a standard cylindrical tip [11] . Our prototype catheter (PC) was developed on this basis. In the first part of the present study, this prototype was tested for functionality and effectiveness of the funnel-shaped extension. The second part of the trial, focused on determining the handling and navigability performance of the new catheter. We used the ACI because the funnel-shaped tip of our PC occludes this lumen of about 0,25-0,35 cm [12] totally occludes. In this segment, too, endovascular treatment has become the standard of care for acute ACI occlusions [13] . In the second part of the experiment, the model was projected onto a screen with the aid of a video camera to realistically simulate working with the catheter and to better assess its handling.

Experimental setup
The experimental setup is shown schematically in the figure above ( Figure 2). under the reference numbers AZ G 14-1-093 and AZ 23177 -07 A16 -1-001 AFW and in accordance with the German Animal Welfare Act (Tierschutzgesetz). All applicable international, national and/or institutional guidelines for the care and use of animals were adhered to. We confirm that the study was carried out in compliance with the ARRIVE guidelines.

The Prototype
The Prototype Catheter (PC) has a modified, funnel-shaped flared distal end made of a vessel-protecting, flexible Nitinol stent mesh. The PC is a modified DAC. It runs inside a Guiding Catheter (GC).
The inner and outer diameters of the PC enable the catheter to fit into a GC and is also large enough to accommodate a microcatheter with stent retriever (SR).
In addition to its guiding function, the GC acts as a shell to prevent the self-expanding tip of the prototype from erecting.
The funnel-shaped end of the new catheter has been designed so that the maximum diameter of the opening extension continues cylindrically over a length of approximately 2-3 cm ( Figure 3). This cylindrical lumen has the same flexibility and erectile force as the proximal funnel-shaped portion. The design was chosen to allow the lumen of the prototype to adapt to the vessel diameter over a longer distance and to provide a protective sheath around the retrievable stent. To ensure visibility under fluoroscopy, the funnel-shaped tip is radiopaque marked.

Thrombectomy procedure
For each thrombectomy procedure, the thrombi were cut in 10 mm pieces and flushed into the arteria carotis interna (ACI) of the vascular model, which was then sealed.
In the first part of the experiment, the PC was introduced into the silicone system in the expanded form, without a guide catheter. During the second part of the experiment, the PC was introduced inside a guiding catheter (GC) (AXS 7 DAC, Catalyst from Stryker). The GC serves as a shell for the funnel-shaped, not yet expanded catheter tip. The microcatheter and a microwire were probed past the thrombus. After removal of the microwire wire, the stent retriever was also released over the thrombus and the funnel-shaped tip was expanded by pulling back the GC. Subsequently, the stent-retriever was deployed for three minutes and then pulled back into the funnel-shaped tip of the PC.

Statistical Analyses
Statistical analyses were performed using SPSS Software (23.0) (IBM, New York, USA). Usually, the chi-squared test is suited for significance tests of independence in contingency tables. However, a prerequisite is that the number of observations in all groups must exceed five, which is not the case for this dataset. Alternatives to the chisquared test include the Fisher's exact test and the Barnard test, which both do not require five observations per group.
Generally, the amount of observations in this dataset is very low (n=20), accordingly, all statistical calculations should be considered carefully.
The level of significance was set at α = 0.05.

First experiment
MT was performed 20 times for each catheter system.  The Null hypothesis (catheter type has no effect on the outcome) cannot be rejected on a 5 % significance niveau (p>0.05). Accordingly, there is no statistically significant improvement in using the new prototype catheter over the old DAC catheter.
The odds ratio provides a measurement of how the odds for successful thrombectomies compare for the two catheters. The results of the Fisher's exact test show an odd´s ratio of 6.1, which means that it is 6.1 times more likely to have a successful thrombectomy using the prototype compared to the DAC. Unfortunately the result is not significant. This is primarily an effect from the small sample size. Repeating the calculations with n = 100 and equal success rates reveals a significantly higher success rate for the prototype catheter.

Second experiment
In 10 of 11 cases, MT was successful in the first pass ( Figure 6). In 1 of 11 cases, MT was only successful in the second pass (Table 2).

Discussion
The aim of the present study is to reduce the risk of thrombus fragmentation during mechanical thrombectomy. For this purpose, a novel catheter was developed and manufactured. In an in vitro setup the prototype was tested for effectiveness and handling. In two different experimental approaches, the performance of the newlydeveloped prototype was tested and compared with a standard DAC.
In the first part of the experiments, successful first-pass recanalization was achieved in 19 out of 20 (95%) cases for the prototype catheter (Figure 7). The thrombus was lost once during retraction before reaching the catheter tip. This was caused by insufficient interaction between the stent retriever and the thrombus. In contrast, MT using the cylindrically shaped commercially available DAC, was successful in 15 of 20 (75%) cases after the first maneuver. The main failure cause was clot abruption and fragmentation at the catheter wall during stent retriever retraction (Figure 8).
Since the thrombus is disproportionally larger than the diameter of common cylindrical catheters, there is a risk that it will be abraded against the catheter wall. We have now been able to confirm the superiority of the funnel-shaped extension from previously published preliminary tests with the new prototype catheter [11] .
In the last 20 years, thrombectomy devices and techniques have evolved significantly with results improving gradually over time [14] . Local antegrade/proximal flow arrest can be achieved by using a BGC (balloon-guided-catheter) to prevent thrombus fragments from being dissipated. As a result, significantly higher recanalization rates, 78.9% using a BCG vs. 67.0% for non-BCG MT (TICI ≤ 2 ); and lower mortality rates were achieved [1] . However, local clot fragmentation and dissipation still remains unresolved.
Another concept known as Larzarus Cover and first described in the scientific literature in 2016, involves covering the stent retriever to reduce embolism in new territories [15] .
The Lazarus Cover consists of a conically-shaped wire mesh sheath that completely encases the stent retriever. It is inserted into the vascular system after placement of the stent retriever and removal of the microcatheter. Using this technique, a recanalization rate of TICI 2b/3 has been achieved in 95% of cases. Evidence of occlusion in new territories could not be observed [16,17] .
The Advanced Thrombectomy System (ANCD) is another novel stroke thrombectomy device featuring a self-expanding funnel-shaped opening, reaching recanalization rates up to 94%. The results of the underlying study stand in accordance with the results achieved by the ANCD (94%) [18] . This emphasizes the possibility of increasing first-pass recanalization rates by using a funnel-shaped distal catheter end.
The main advantages of the newly-developed catheter tested in this study are 1) the prevention of clot abruption and fragmentation at the catheter wall enabled by the flexible and funnel-shaped tip, 2) inducing local flow arrest by "sealing" the vessel lumen after deployment of the dilated end, and 3) ensuring safe retraction by encasing.
The second part of the experiments registered a first-pass recanalization rate of 91%.
A fragment of the thrombus was lost before reaching the prototype in 1 out of 11 passes due to insufficient interaction between the thrombus and the stent retriever. The navigability of the new prototype was assessed in an in vitro vessel model by using a video camera, and a monitor to simulate an angiography setting. MT were performed by an experienced interventionalist, and both procedure and materials were consistent with routine practice. Handling and navigability of the prototype were not impaired.
Thus, the present study confirms the results of previous in vitro studies published by our research group. Moreover, the prototype catheter with its flexible and self-erecting end successfully demonstrated improved recanalization rates compared with conventional standard DACs. The new prototype has the potential to improve first-pass recanalization rates and to reduce clot migration into distal territories.

Conclusion
The experimental series demonstrated that the newly-developed prototype catheter featuring a funnel-shaped and self-expanding tip achieved a higher first pass recanalization rate than a cylindrical standard DAC.           The exible tip of the prototype catheter is 2.4 cm long and consists of a exible nitinol mesh.  The prototype catheter achieved a success rate of 95%. The DAC achieves a success rate of 75%.

Figure 6
Success of mechanical thrombectomy with the prototype catheter in the rst pass Thrombectomy under uoroscopy with the prototype catheter (arrow) and stent retriever. The funnelshaped expanded tip of the prototype lls the vessel lumen, enalbing the thrombus to be completely retracted into the DAC. A and B show retraction of the stent retriever toward the prototype catheter. C and E show retraction of the stent retriever into the prototype catheter. F shows retraction of the prototype catheter with trapped stent retriever and thrombus, with no evidence of thrombus fragmentation and migration.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. Supplementaryinformation.mov