Article Text
Abstract
Introduction/Purpose Neurovascular catheters can be used in numerous ways to treat ischemic stroke. These devices have potential to cause vessel injury by disrupting the endothelium during treatment. Although vessel injury is typically evaluated in animal models, prior work suggests that endothelialized in vitro models may be developed as an additional option in the preclinical testing pipeline. The purpose of the current work was to create a complex, clinically-relevant neurovascular silicone model that is fully endothelialized, and to determine the utility of this model for evaluating vessel injury due to catheter simulated-use.
Materials and Methods Complex anatomical models that represented the ICA and MCA were fabricated out of silicone, with an inner diameter tapering from 5.5mm ID to <4mm ID and a total tracking length of 256.3mm. These silicone models were sterilized, rinsed, and coated with fibronectin. Fibronectin-coated silicone models were placed in individual customized bioreactor systems and seeded with human umbilical vein endothelial cells. After cell injection, vessels were maintained on peristaltic pumps and housed in a large incubator. An initial ‘model characterization’ study was performed to determine if the complex models could be successfully endothelialized and maintained. For this study, vessels were cultivated under flow for 3 or 7 days, then harvested and analyzed using H&E for cell morphology and PECAM/BBI for cell quantification and phenotype confirmation. Following the characterization study, ‘catheter simulated-use’ studies were performed. For these studies, vessels were cultivated for 3 days, then treated with a guidewire alone, ‘biaxial’ approach (guidewire plus microcatheter), or ‘triaxial’ approach (guidewire plus microcatheter plus aspiration catheter, and treatments were either performed one time (1-pass) or three times (3-pass). There were n=3 vessels per treatment and number of passes, plus controls, for a total of n=24 vessels. Immediately following treatment, vessels were harvested and fixed, then ready for analysis. Vessels were stained with H&E and evaluated for vessel injury, using a quantitative ImageJ approach as well as a visual scoring system. Vessels were stained with PECAM to confirm endothelial cell phenotype.
Results H&E images from vessels in the characterization study revealed a consistent and confluent monolayer of endothelial cells throughout the length of the vessels at both 3 and 7 days. PECAM/BBI results further illustrated the expected cell morphology, while nuclei quantification illustrated cellular coverage at densities consistent with native endothelium. Results from the catheter simulated-use studies demonstrated that vessel injury can be successfully visualized and characterized using quantitative and visual scoring methods. Results specifically revealed that injury increased with the number and size of devices and with the number of passes. The differences and trends were consistent with expected outcomes.
Conclusion This work demonstrates that complex, clinically-relevant silicone models can be successfully endothelialized and used to visualize and assess injury following catheter simulated-use. Results were consistent with expected trends based on number of passes and treatment approach, which supports future use of the model to evaluate and compare a variety of neurovascular devices. Overall, this provides a cost-effective, early-stage in vitro model that can precede or complement standard animal testing.
Disclosures A. McCulloch: 1; C; This work was funded by Stryker Neurovascular. B. Yang: 1; C; This work was funded by Stryker Neurovascular. S. Frenklakh: 5; C; Employee of Stryker Neurovascular. P. Sah: 5; C; Employee of Stryker Neurovascular. K. Cardinal: 1; C; This work was funded by Stryker Neurovascular.