Implanted cardiovascular devices must undergo a range of applied cyclic loads after implantation. Peripheral devices, for example, are subjected to both radial pulsatile loading and non-radial loading such as cyclic bending from limb motion. Coronary stents may also see a range of loads due to increases in heart exertion, such as during exercise. As highlighted by these examples, such variable loading commonly involves a smaller number of high amplitude cycles combined with a larger number of low amplitude cycles.

Typical S/N type testing to determine the fatigue limit of Nitinol is done over a range of strain amplitudes; however, each specimen sees only a single peak strain amplitude during cycling. The effect of variable loading on Nitinol is therefore not understood. The purpose of this study was to evaluate any potential cumulative fatigue effect of combining low strain amplitude cycles with high strain amplitudes cycles on Nitinol wire apex specimens. A series of fatigue tests were performed to evaluate the fatigue response of Nitinol to variable loading. The results demonstrated that fatigue cycles at lower strain amplitudes can limit the number of higher amplitude cycles to failure in a variable loading scenario. However, the results also indicate that a small number of higher amplitude cycles can dominate the fatigue damage; almost all fractures occurred shortly after completing a section of higher amplitude cycles.