Determination of coefficient of friction for self-expanding stent-grafts.

Migration of stent-grafts (SGs) after endovascular aneurysm repair of abdominal aortic aneurysms is a serious complication that may require secondary intervention. Experimental, analytical, and computational studies have been carried out in the past to understand the factors responsible for migration. In an experimental setting, it can be very challenging to correctly capture and understand the interaction between a SG and an artery. Quantities such as coefficient of friction (COF) and contact pressures that characterize this interaction are difficult to measure using an experimental approach. This behavior can be investigated with good accuracy using finite element modeling. Although finite element models are able to incorporate frictional behavior of SGs, the absence of reliable values of coefficient of friction make these simulations unreliable. The aim of this paper is to demonstrate a method for determining the coefficients of friction of a self-expanding endovascular stent-graft. The methodology is demonstrated by considering three commercially available self-expanding SGs, labeled as A, B, and C. The SGs were compressed, expanded, and pulled out of polymeric cylinders of varying diameters and the pullout force was recorded in each case. The SG geometries were recreated using computer-aided design modeling and the entire experiment was simulated in ABAQUS 6.8/STANDARD. An optimization procedure was carried out for each SG oversize configuration to determine the COF that generated a frictional force corresponding to that measured in the experiment. The experimental pullout force and analytically determined COF for SGs A, B, and C were in the range of 6-9 N, 3-12 N, and 3-9 N and 0.08-0.16, 0.22-0.46, and 0.012-0.018, respectively. The computational model predicted COFs in the range of 0.00025-0.0055, 0.025-0.07, and 0.00025-0.006 for SGs A, B, and C, respectively. Our results suggest that for SGs A and B, which are exoskeleton based devices, the pullout forces increase upto a particular oversize beyond which they plateau, while pullout forces showed a continuous increase with oversize for SG C, which is an endoskeleton based device. The COF decreased with oversizing for both types of SGs. The proposed methodology will be useful for determining the COF between self-expanding stent-grafts from pullout tests on human arterial tissue.