Mathematical and Computational Modeling of Metal Ion Release from Cardiovascular Implants
Date Issued
May 2024
Author(s)
Advisor
Abstract
In biomedical engineering, ensuring the safety and efficacy of implants through rigorous
testing procedures is paramount. The majority of metal alloys used for cardiovascular
implants contain high levels of nickel, which, if released in sufficient quantities, can lead
to several adverse local as well as systemic effects. To assess general corrosion
susceptibility and metal ion release, the US Food and Drug Administration (FDA)
recommendations include testing per the American Standard of Testing and Materials
(ASTM) F2129-08, F3306-19, and G31-72(2004). Modeling and simulation tools
represent a promising approach to improve the nonclinical testing methods employed in
the Toxicological Risk Assessment (TRA) of medical devices.
This thesis utilizes physics-based equations, stochastic modeling, and computation
simulations to complement and supplement the use of in vitro experiments. Physics-based
equations will be used as prediction tools on current reported data, which include varying
post-processing techniques, varying chemical conditions, application of mechanical
stress, and different fabrication methods. Stochastic modeling (Monte-Carlo probabilistic
simulations) will address how data are reported in these protocols and their possible use
in predictive models. Lastly, computational simulations will be developed to obtain
diffusion coefficients and concentration of nickel as these are difficult to measure
experimentally. Applying the above shows that post-processing techniques and external
factors, such as stress and chemical conditions of immersion, influence nickel ion release.
This is evident by the parameters and data fitting predictions obtained by the tested
physics-based equations. Parameters are then incorporated into the Monte Carlo
simulations to be used as inputs that consist of a range of values, with the ultimate goal
of assisting in better nickel release predictions when combined with larger and more
complex predictive models. Lastly, computational simulations have proven useful in
providing approximate values of diffusion coefficients of nickel in the surrounding
medium and surface Ni concentration, both of which are critical to nickel ion release.
testing procedures is paramount. The majority of metal alloys used for cardiovascular
implants contain high levels of nickel, which, if released in sufficient quantities, can lead
to several adverse local as well as systemic effects. To assess general corrosion
susceptibility and metal ion release, the US Food and Drug Administration (FDA)
recommendations include testing per the American Standard of Testing and Materials
(ASTM) F2129-08, F3306-19, and G31-72(2004). Modeling and simulation tools
represent a promising approach to improve the nonclinical testing methods employed in
the Toxicological Risk Assessment (TRA) of medical devices.
This thesis utilizes physics-based equations, stochastic modeling, and computation
simulations to complement and supplement the use of in vitro experiments. Physics-based
equations will be used as prediction tools on current reported data, which include varying
post-processing techniques, varying chemical conditions, application of mechanical
stress, and different fabrication methods. Stochastic modeling (Monte-Carlo probabilistic
simulations) will address how data are reported in these protocols and their possible use
in predictive models. Lastly, computational simulations will be developed to obtain
diffusion coefficients and concentration of nickel as these are difficult to measure
experimentally. Applying the above shows that post-processing techniques and external
factors, such as stress and chemical conditions of immersion, influence nickel ion release.
This is evident by the parameters and data fitting predictions obtained by the tested
physics-based equations. Parameters are then incorporated into the Monte Carlo
simulations to be used as inputs that consist of a range of values, with the ultimate goal
of assisting in better nickel release predictions when combined with larger and more
complex predictive models. Lastly, computational simulations have proven useful in
providing approximate values of diffusion coefficients of nickel in the surrounding
medium and surface Ni concentration, both of which are critical to nickel ion release.
Subjects