A computational framework to optimize the mechanical behavior of synthetic vascular grafts

Abstract

The failure of synthetic small-diameter vascular grafts has been attributed to a mismatch in the compliance between the graft and native artery, driving mechanisms that promote thrombosis and neointimal hyperplasia. Additionally, the buckling of grafts results in large deformations that can lead to device failure. Although design features can be added to lessen the buckling potential (e.g., reinforcing coil), the addition is detrimental to decreasing compliance. Herein, we developed a novel finite element (FE) framework to inform vascular graft design by evaluating compliance and resistance to buckling. A batch-processing scheme iterated across the multi-dimensional design parameter space, which included three parameters: coil thickness, modulus, and spacing – generating 100 unique designs. FE models were created for each coil-reinforced graft design to simulate pressurization, axial buckling, and bent buckling, and results were analyzed to quantify compliance, buckling load, and kink radius, respectively. Validation of the FE models demonstrated that model predictions agreed with experimental observations for compliance ($r$ = 0.99), buckling load ($r$ = 0.89), and kink resistance ($r$ = 0.97). Model predictions demonstrated a broad range of values for compliance (1.1–7.9 %/mmHg × 10⁻²), buckling load (0.28–0.84 N), and kink radius (6–10 mm) across the design parameter space. Subsequently, data for each design parameter combination were optimized (i.e., minimized) to identify candidate graft designs with promising mechanical properties. Our model-directed framework successfully elucidated the complex mechanical determinants of graft performance, established structure-property relationships, and identified vascular graft designs with optimal mechanical properties, potentially improving clinical outcomes by addressing device failure.