Improving the hemocompatibility of a porohyperelastic layered vascular graft using luminal reversal microflows

Abstract

Vascular graft thrombosis is a long-standing clinical problem. A myriad of efforts have been devoted to reducing thrombus formation following bypass surgery. Researchers have primarily taken a chemical approach to engineer and modify surfaces, seeking to make them more suitable for blood contacting applications. Using mechanical forces and surface topology to prevent thrombus formation has recently gained more attention. In this study, we have designed a bilayered porous vascular graft capable of repelling platelets and destabilizing absorbed protein layers from the luminal surface. During systole, fluid penetrates through the graft wall and is subsequently ejected from the wall into the luminal space (Luminal Reversal Flow - LRF), pushing platelets away from the surface during diastole. In-vitro hemocompatibility tests were conducted to compare platelet deposition in high LRF grafts with low LRF grafts. Graft material properties were determined and utilized in a porohyperelastic (PHE) finite element model to computationally predict the LRF generation in each graft type. Hemocompatibility testing showed significantly lower platelet deposition values in high versus low LRF generating grafts (median±IQR = 5,708 ± 987 and 23,039 ± 3,310 platelets per mm², respectively, p=0.032). SEM imaging of the luminal surface of both graft types confirmed the quantitative blood test results. The computational simulations of high and low LRF generating grafts resulted in LRF values of -10.06 μm/s and -2.87 μm/s, respectively. These analyses show that a 250% increase in LRF is associated with a 75.2% decrease in platelet deposition. PHE vascular grafts with high LRF have the potential to improve anti-thrombogenicity and reduce thrombus-related post-procedure complications. Additional research is required to overcome the limitations of current graft fabrication technologies that further enhance LRF generation.

Keywords: Hemocompatibility; Luminal reversal flow; Polymer biomaterial; Porohyperelastic; Vascular graft.

Fig. 1.

Cross-section of the bilayered PHE graft to demonstrate Luminal Reversal Flow (LRF) using velocity vectors.

Fig. 2.

Tubular biaxial tensile testing device used to mechanically characterize the single-layered constructs in the circumferential direction. The chamber was heated to 37 °C. The construct inlet was connected to the pressure line and the outlet was blocked. A camera was used to monitor the construct diameter (red circle tracking in figure). An average of these diameters on the last cycle were used in our analysis.

Fig. 3.

Representative SEM images of the electrospun (a) Tecoflex and (b) Thermobonded PCL. Thermobonded PCL image was adapted from Behrangzade et al. (2023). These images were processed in MATLAB to determine the porosity for the void-dependent permeability calculations in the PHE simulations.

Fig. 4.

In-vitro mock flow loop used to perform hemocompatibility tests. Fresh citrated whole ovine blood was circulated in a mock flow loop for 90 min under pulsatile flow. A pressure transducer (PT) was used to monitor the pressure range.

Fig. 5.

(a) Mechanical responses of Tecoflex SG-85 A and thermobonded PCL (each n = 3) in the circumferential direction. (b) Tangential moduli of both materials at the maximum stretch ratio. The deformation of TPCL constructs was significantly lower than Tecoflex constructs in response to the same pressurization range (p = 0.039). No axial displacement was imposed during these tests, however, the axial loads measured during inflation were all below 0.02N.

Fig. 6.

Objective function values in the computational DOE domain for Tecoflex SG-85 A (a) and thermobonded PCL (b). Green points indicate the minimum of the domain based on the objective function values (Eq. (6)). These data points represent each material’s properties and were used to calculate the LRF in each graft type.

Fig. 7.

Representative SEM images of high LRF (a) and low LRF (b) showing the permeability and thickness of the graft layers.

Fig. 8.

LDH assay results for platelet deposition per luminal surface area and computational predictions of LRF in each graft type using the PHE FE model.

Fig. 9.

Representative SEM images of high LRF (a–b) and low LRF (c–d) luminal surfaces exposed to citrated ovine whole blood for 90 min under pulsatile flow. Higher LRF grafts had lower platelet deposition compared to the grafts with lower LRF. The SEM images confirm LDH assay results.