Development of physiologically relevant synthetic thrombus for use in visual analysis of in vitro mechanical thrombectomy device testing

Abstract

Background:

Ischemic stroke is a leading cause of death and significant long-term disability worldwide. Mechanical thrombectomy is emerging as a standard treatment for eligible patients. As clinical implementation of stent retrieval and aspiration thrombectomy increases, a need exists for physiologically relevant in vitro device efficacy testing. Critical to this testing is the development of standardized “soft” and “hard” synthetic blood clots that mimic the properties of human thrombi and are compatible with imaging technologies. Synthetic clots allow researchers to extract information regarding clot integration, model hemodynamics, and quantify the physics of thrombectomy.

Methods

This work develops polyacrylamide and alginate-based synthetic clots that are compatible with particle image velocity (PIV) and radiographic imaging techniques while maintaining mechanical properties of “soft” and “hard” human clots. Dynamic mechanical analysis (DMA) testing using an HR2-Rheometer demonstrates comparable mechanical properties to human clots previously tested by this research group and provided in existing literature.

Results

The synthetic clots are formulated with either 0.5% w/v polyethylene microspheres for PIV visualization or 20% w/v barium sulfate for angiographic visualization, enabling real-time imaging of clot behavior during thrombectomy simulations. The soft formulation shows compressive and shear properties of 8–9 KPa and 2–3 KPa, respectively. The hard clots are 4–6x stiffer, with compressive and shear properties of 41–42 KPa and ~12 KPa, respectively.

Conclusion

Standardized synthetic clots offer a platform for reproducible device testing. This provides a greater understanding of mechanical thrombectomy device efficacy, which may lead to quantifiable advances in device development and eventual improved clinical outcomes.

Introduction

Stroke is a leading cause of death and serious long-term disability worldwide. Ischemic stroke is the most common, and its global health burden is steadily increasing with 9.62 million cases and nearly 5 million deaths predicted in the year 2030.^1,2 Mechanical thrombectomy via an aspiration catheter is emerging as the standard of care for eligible patients with large vessel occlusion (LVO) stroke over stent retrievers or drug therapy via tissue plasminogen activator (tPA).^3–5

The widespread adoption of mechanical thrombectomy necessitates robust in vitro testing strategies to reliably evaluate the effectiveness of aspiration catheter devices. An optimal strategy for aspiration catheter testing includes a sophisticated in vitro neurovascular vessel model and synthetic, physiologically relevant blood clots. Although there have been marked developments in benchtop models, research institutions and industry lack synthetic clots with quantified, reproducible, and adjustable mechanical properties that are compatible with the latest in imaging technologies.

Commercially available animal-based and synthetic clots offer partial solutions but lack standardized mechanical properties that are necessary to reduce comparison variability in device testing. Animal-based clots can simulate the physiological variations seen in human thrombus, such as red blood cell (RBC) content, fibrin content, and calcifications. However, these animal-based clots vary widely in size and mechanical properties.^6–9 Commercially available synthetic clots, such as those offered by BioModex (Quincy, MA), are useful for demonstrations and surgical simulations but do not have quantified mechanical properties that are statistically similar to known human blood clot classifications.^10,11 Modifications to polymer-based synthetic clots (previously developed by this research group) provide a reliable analog that is visualizable with PIV or angiography, formulated to specific and repeatable mechanical properties, non-biohazardous, and stable with a longer shelf-life.¹²

Numerous studies have shown that clot composition affects aspiration catheter efficacy and patient outcomes after mechanical thrombectomy.^13–20 To assess device efficacy in vitro models, there is a need for synthetic clots that have standardized mechanical properties that cover a range of human clot classifications. We propose two polymer-based synthetic clot formulations with standardized and repeatable mechanical properties that mimic the human thrombi classifications of “soft” and “hard” clots. These two formulations represent 80% of the human clots removed via mechanical thrombectomy each year.^12,20,21 Categorizing thrombi based on mechanical properties, specifically “soft” or “hard”, provides a framework for tailoring in vitro neurointerventional studies to match clinical scenarios.

Human blood clots are not radiopaque and must be identified clinically by the lack of flow in a neurovascular branch. Synthetic blood clots compatible with the latest imaging technologies provide quantifiable information about clot integration and aspiration thrombectomy kinetics previously unavailable in the clinical setting. Particle Image Velocimetry (PIV) and angiographic imaging are useful techniques for clot visualization and understanding clot kinetics. PIV, a technique popularized to measure complex fluid fields, can help better understand the aspiration forces, velocities, and clot compressive and shear effects during aspiration.^22–24 High frame-rate angiographic imaging can provide additional information on clot position in the vessel, clot interaction at the catheter tip (clot engagement), clot incorporation into the aspiration catheter (clot ingestion), and clot migration through the catheter (clot aspiration) during vacuum pump or syringe aspiration.

In this work, reproducible “soft” and “hard” clots are formulated to be visible via PIV and angiography imaging. The PIV synthetic clots contain fluorescent polymer microspheres for visualization with a UV light source during simulated surgical demonstrations and training. Discrete PIV clot ingestions can be captured in real time using a laser source to highlight the clot and a high-speed camera to capture the clot aspiration. PIV video analysis provides quantifiable flow velocities, stresses, strains, and force interactions between the catheter tip and the clot. The radiopaque synthetic clots contain barium-sulfate for visualization via angiography to provide clinically relevant in vitro visualization of the clot during full benchtop surgical simulations.

Improvements in synthetic clots, compatible with the latest imaging technology, can serve as useful tools for aspiration catheter testing, comparison, and selection, and may ultimately help guide clinical decision making.

Methods

This work summarizes the creation, validation, and imaging of synthetic clot formulations with repeatable mechanical properties matching “soft” and “hard” human thrombi. PIV-compatible clots containing polyethylene microspheres and radiopaque clots containing barium sulfate were both created with “soft” and “hard” properties for use during surgical simulations and thrombectomy device testing. The PIV-visible and radiopaque synthetic clots were iterations of the previously published work from this research group on polyacrylamide and alginate (PAAM-Alg) synthetic blood clots.¹² Clot material properties were adjusted by modifying solvent percentages during the clot formulation process, in order to match human thrombus mechanical properties described in literature.^12,15,17,21 “Soft” and “hard” clot properties were validated by testing the compressive stiffness (elastic complex modulus – E*) and shear effects (shear complex modulus – G*) using a dynamic mechanical analysis (DMA) HR2-Rheometer (TA Instruments, New Castle, DE). Clot visualization was verified using both in-house PIV and angiographic imaging techniques.

Clot creation and formulation

Human thrombi are typically classified in qualitative terms— red blood cell (RBC)-rich (soft), or fibrin-rich (hard). However, recent literature provides more quantitative definitions of clot types based on composition, with soft clots containing ~60% RBCs and hard clots containing ~20% RBCs.^15,17 Previous mechanical testing of human RBC-rich (soft) clots resulted in an average elastic modulus of 12.8 kPa, while fibrin-rich (hard) clots have a compressive modulus approximately three to four times greater than soft clots (36–48 kPa).^12,21 These benchmark values guided this work, which iterated upon recent PAAM-Alg clot formulations to further develop “soft” and “hard” clots with repeatable mechanical properties. Physiologically-relevant synthetic clots were made by combining polyacrylamide and alginate (PAAM-Alg) with a crosslinking agent per the research group’s previously published procedure.¹² In summary, an aqueous solution was prepared with either 20% w/v barium sulfate (BaSO₄ for radiopacity, Sigma-Aldrich, St. Louis, MO)) or 0.5% w/v of fluorescent polyethylene microspheres suspended in a surfactant solution (for PIV imaging, ~50 μm diameter, 300–600 nm fluorescent wavelength, Cospheric LLC, Somis, CA). Sodium alginate and acrylamide were then added and stirred until fully dissolved, followed by the referenced cross-linking agents.¹² The gelling fluid was then aspirated into thin-walled stainless-steel tubing that matched specific vessel dimensions (2–5mm internal diameter (ID)). The gels were left to cure in an incubator (40–50°C) for 18 hours. For clots made with PIV particles, the tubes were slowly rotated (1 rpm) during the curing process to ensure even particle distribution in the final clot.

Clot mechanical testing – shear and compression

To validate the mechanical properties of the synthetic clots, a non-destructive testing procedure was performed with the DMA-Rheometer. Synthetic clot samples were molded into 8 mm (diameter) by 4 mm (height) cylinders. Samples were then tested with an 8 mm diameter head geometry and temperature-controlled quick-exchange Peltier plate base, both fixed with 150-grit adhesive sandpaper to reduce sample slippage.

To measure compressive (E*) and shear (G*) moduli, each sample was preloaded to 0.2 N force and tested over a physiologically relevant frequency sweep. E* was measured with a head geometry oscillation of ± 20 μm across a frequency range from 0.1 to 10 rad/s. G* was measured with a 1% rotational shear strain applied and released across the same frequency range from 0.1 to 10 rad/s. The tests were repeated three (n=3) times for each sample. The samples were kept hydrated with deionized water throughout all mechanical tests. All tests were performed at room temperature (20°C) to mimic standard in vitro device testing conditions. Synthetic clot mechanical properties at room temperature were compared to human clot data at body temperature (37°C) to bridge the gap between a room temperature in vitro model and physiological relevance.

Clot visualization – PIV particles

Particle image velocimetry (PIV) is a visualization tool for analyzing and tracking the movement (velocity) and kinetic interactions (force, stress, and strain) of a system of interest that contains fluorescent PIV particles. PIV particles embedded in synthetic clots or mixed in a blood analog allow for real-time analysis of clot ingestion and aspiration catheter interactions. PIV particle-containing clots were visualized with a 532 μm “green” laser and recorded with a high-speed camera.

Clot visualization – angiography

Angiography with video recording of a radiopaque clot provides information regarding clot aspiration times and ingestion behaviors during mechanical thrombectomy. Barium sulfate-containing clots were placed inside of a comprehensive circle of Willis (CW) benchtop model and visualized using fluoroscopy-guided angiography.

Results

Clot mechanical properties – shear and compression

This work demonstrated that variations in solvent wt% (deionized water) and barium sulfate powder can independently alter the shear and elastic modulus. Clots with 20% w/v barium sulfate exhibited stiffer mechanical properties than the clots with PIV particles. To standardize a physiologically similar “hard” and “soft” clot, adjustments were made to the weight percentage of the solvent while keeping the weight percentage of PAAM-Alg constant for both clots (93 wt%). Clot formulations were made by adjusting the solvent percentage from 80% - 89% for PIV clots and 87% - 93% for barium sulfate clots (Figure 1).

Figure 1:

Compressive mechanical properties (elastic modulus) of PIV clots and barium sulfate clots versus solvent content.

Hard clots had repeatable properties using 82% solvent for the PIV clots and 87% solvent for barium sulfate clots. Soft clots had repeatable properties using 89% solvent for the PIV clots and 93% solvent for barium sulfate clots (Table 1).

Table 1:

Mechanical testing results - shear and compression data for synthetic clots containing particles for PIV imaging and barium sulfate for radiographic imaging

Clot Type	Shear Modulus (G* - kPa)	Standard Deviation	Compressive Modulus (E* - kPa)	Standard Deviation
Soft Clot (93% PAAM, 93% Solvent with BaSo4)	2.1	12.7%	12.2	11.5%
Soft Clot (93% PAAM, 89% Solvent, with PIV Particles)	2.9	7.2%	12.3	16.2%
Hard Clot (93% PAAM, 87% Solvent with BaSo4)	8.0	5.6%	40.8	7.9%
Hard Clot (93% PAAM, 82% Solvent, PIV Particles)	9.3	3.1%	41.7	4.6%

For shear, the mechanical properties of barium sulfate clots were 2.1 kPa for soft clots and 8.0 kPa for hard clots. The PIV clots had statistically similar mechanical properties, averaging 2.9 kPa for soft clots and 9.3 kPa for hard clots.

For compression, the mechanical properties of barium sulfate clots averaged 12.2 kPa for soft clots and 40.8 kPa for hard clots. The PIV clots had statistically similar mechanical properties, averaging 12.3 kPa for soft clots and 41.7 kPa for hard clots.

Clot visualization verification – PIV particles

PIV clot imaging was verified by illuminating the clots with a 350 μm UV light source and placing the clots in a discrete and transparent vessel model in contact with an aspiration catheter. The setup allowed for tracking the clot during placement, benchtop testing and surgical simulations. Laser fluorescence with a high-speed camera captured PIV clot ingestion into an aspiration catheter tip during a simulated mechanical thrombectomy procedure (Figure 2).

Figure 2-.

top: PIV clot fluoresced with a 350 μm UV light source, bottom: PIV clot ingestion into an aspiration catheter tip.

Future studies will utilize a similar high-speed camera and laser fluorescence setup to visualize PIV clot ingestions during comprehensive real-time thrombectomy simulations. The resulting data can be used to create velocity vector maps, calculate strain rates, and quantify stresses and forces during clot ingestion.

Clot visualization verification – angiography

Radiographic clot imaging was verified by placing clots within a comprehensive circle of Willis (CW) benchtop model (Figure 3). Utilizing endovascular surgical techniques, an aspiration catheter was engaged with the clot in the right internal carotid artery (RICA) of the model and aspirated under digital subtraction angiography (DSA) using a C-arm fluoroscope (Skanray Scan-C, Romeo, MI).

Figure 3:

A – fluoroscopic image of a circle of Willis (CW) model, B - barium sulfate clot placement in right internal carotid (RICA) under digital subtraction (DSA), C – aspiration catheter engagement with clot (DSA), D – fluoroscopic image of partial ingestion: clot proximal end compressed and stretched inside the catheter, with clot distal end still outside the catheter tip (DSA).

Future studies will utilize radiographic clots of repeatable dimensions and mechanical properties to test the aspiration efficiency of various aspiration catheters under simulated surgical conditions.

Discussion

In this work, synthetic clots were created with quantified physiologically relevant mechanical properties corresponding to “soft” and “hard” human clots in standardized shapes and sizes. These synthetic clots have the added benefit of being compatible with PIV and angiography which allows for qualitative and quantitative assessments of mechanical thrombectomy procedures and statistical comparison of treatment device effectiveness. The incorporation of PIV particles enables a detailed analysis of clot movement, velocity, and aspiration forces. Radiopaque synthetic clots, visible via fluoroscopy and angiography, provide additional information during aspiration, such as clot engagement, ingestion, and migration through the catheter. The creation and mechanical validation of reproducible polymer-based synthetic clots fill a critical gap in device testing capabilities, informing future advancements in neurointerventional device development.

Conclusions

Synthetic clots that are compatible with modern imaging modalities present opportunities in data collection that can inform neurointerventional approaches and tailor strategies based on clot classifications. Future research directions involve refining clot formulations for other clot types (i.e calcified, fibrin rich, etc.), exploring additional imaging technologies, and further enhancing the relevance of synthetic clots for device testing. These research findings promote further research and development in stroke treatment by offering quantifiable results, leading to a more nuanced understanding of thrombi and aspiration device behavior during thrombectomy.

Key Messages.

What is already known on this topic:

Synthetic and animal blood clots are well established for device demonstrations and surgical trainings, however, clots with consistent and validated mechanical properties that match the human condition are not readily available.

What this study adds:

These synthetic clots are engineered to match physiological material properties of human clots and have standardized and repeatable mechanical properties. The clots also include novel Particle Imaging Velocimetry (PIV) and contrast agents for compatibility with advanced PIV and angiographic imaging technologies.

How this study might affect research, practice, or policy:

The clots can provide improved realism to in vitro surgical simulations and quantify statistical assessment of medical device performance. Mechanically matched “hard” and “soft” can help promote further research and development into advancing stroke treatment devices by offering quantifiable results, leading to a better understanding of thrombi and aspiration device behavior during thrombectomy.