Fatigue Life Prediction Methodology for Superelastic Nitinol Implantable Medical Devices

Predicting the fatigue life of Nitinol implantable medical devices presents significant challenges. The absence of an accurate constitutive model that captures the nonlinear superelastic behavior, loading and unloading hysteresis, and the nonlinear strain-hardening after plastic deformation, all complicate efforts to characterize the true stress/strain state in test samples and devices. For superelastic materials like Nitinol, the path-dependent hysteresis during phase transformation makes stress-based fatigue prediction difficult. To address this, strain-based techniques have traditionally been used to estimate Nitinol’s fatigue life. However, prior research has shown inconsistencies between strain-based finite element simulations and actual strain measurements in the mixed-phase regime using digital image correlation. The present study re-examines the classic stress-based fatigue analysis for Nitinol devices in an effort to eliminate these inconsistencies. We investigated applying a stress-based methodology to the very high cycle fatigue analysis of Nitinol devices using the superposition of the residual stress from pre-conditioning and the applied in-vivo stress as the key parameter for estimating fatigue life under various conditions. Multiple residual stress characterization techniques were explored to validate the stress-based finite element predictions. The classical constant-life diagram in the stress domain was revisited and compared to the traditional strain-domain approach. The stress-based methodology appears to be suitable for assessing the very high cycle fatigue reliability of superelastic Nitinol long-term implantable medical devices requiring high reliability and confidence.