Understanding the Deformation and Fracture of Nitinol Endovascular Stents Using In Situ Synchrotron X‐Ray Microdiffraction

Abstract

Endovascular stents manufactured from superelastic Nitinol represent a major component in the fight against heart disease. However, accurate characterization of the stress/ strain distributions in such stents, which govern their deformation and fracture behavior, is essential for their prolonged safe use in human arteries. Nitinol, a nearly equiatomic alloy of nickel and titanium, can “remember” a previous shape and can recover strains as high as 10 % by deformation (superelasticity) or temperature change (shape memory). These properties result from a reversible first-order phase transition between austenite (cubic, B2) and martensite (monoclinic, B19′). As such, deformation mechanisms of Nitinol are more complex than the conventional modes of plastic deformation in traditional alloys. Consequently, the mechanical behavior of Nitinol under multiaxial conditions remains poorly understood. Nevertheless, because of these unique mechanical characteristics, in combination with excellent biocompatibility, Nitinol is used as self-expanding endovascular stents to scaffold diseased peripheral arteries. First-generation Nitinol stents were designed to provide sufficient scaffolding forces to hold open vessels, yet provide enough elasticity to “breathe” with pulsatile pressure differentials from the cardiac cycle. A variety of clinical studies indicate that these stents perform this primary function quite well. More recent in-depth studies, however, reveal that superficial femoral arteries (SFAs) are subjected to complex in vivo multiaxial deformation with up to 60 % rotation and ca. 20 % contraction in the SFA as the leg is bent from an extended position. Correspondingly, during a walking cycle, a stent deployed in the SFA undergoes severe multiaxial displacements from pulsatile motion (ca. 4 × 10 cycles annually) plus bending, torsion, and axial motions (at a rate of ca. 1 × 10 cycles annually). Although there are ca. 40 times more cardiac displacement cycles, the combined nonpulsatile motions result in far greater cyclic strain magnitudes, and therefore, have the possibility of inducing greater fatigue damage. Stent design for these “dynamic” arteries and subsequent prediction of conditions and locations of likely fatigue-induced fracture events is invariably performed with detailed finite-element analysis. These numerical models attempt to incorporate the nonlinear mechanical properties of the Nitinol constitutive relationship to provide an estimate of the distributions of local stresses and strains. Nevertheless, endovascular stents still fracture in vivo! Accordingly, we may conclude that either our knowledge of the nature and magnitude of the deformation of a stented artery is incomplete, or current finite-element models fail to fully represent the actual mechanical response of a Nitinol stent. Deformation of a Nitinol stent (e.g., Fig. 1a) due to in vivo loading conditions can be modeled with bending and unbending of the repeating structural “half diamonds”. To simulate these stent features, our test sample consists of a planar object composed of two opposed half diamonds that we refer to as a “diamond specimen” (Fig. 1b). These diamonds contain the salient geometric features of many Nitinol self-expanding stents, and are thus the ideal configuration for fundamental strain and fatigue analyses. The computed eyy components of the deviatoric strain tensor in the diamond from the synchrotron Laue microdiffraction experiments are shown in Figure 2. Under 0 mm displace– [*] Prof. R. O. Ritchie Department of Materials Science & Engineering University of California Berkeley, CA 94720 (USA) E-mail: roritchie@lbl.gov Dr. A. Mehta SSRL/SLAC Stanford University Menlo Park, CA 94025 (USA) Dr. X.-Y. Gong Medical Implant Mechanics, LLC Fremont, CA 94539 (USA) Dr. V. Imbeni SRI International Menlo Park, CA 94025 (USA) Dr. A. R. Pelton Nitinol Devices & Components Fremont, CA 94539 (USA) [**] This work was supported by Nitinol Devices & Components, Inc. (NDC), a Johnson & Johnson Co., by the National Science Foundation under Grant No. CMS-980006 (for R.O.R.), and by the U.S. Department of Energy under Contract No. DE-AC02-76SF00515 (for A.M.). The microdiffraction results were obtained at beamline 7.3.3 at the Advanced Light Source at the Lawrence Berkeley National Laboratory, which is supported by the Director, Office of Science, Office of Basic Energy Sciences, of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. We are very grateful to Dr. Nobumichi Tamura (of ALS/LBNL) for his help and guidance in collecting and processing the finite-element analysis data and to Dr. Brad L. Boyce (Sandia National Laboratory) for designing and fabricating the in situ straining rig. We are also very grateful to Dr. Scott W. Robertson (University of California, Berkeley) and Dr. Monica Barney (NDC) for helping us run the beamline. SLAC-PUB-12394