Medical Devices Home      Exposition      To Register      ASM Homepage
Back to "Materials Modeling" Search
    Back to Main Search

Tuesday, August 11, 2009 - 9:40 AM
4.4

Simulated Tissue-Implant Interaction In a Minimally Invasive Mitral Valve Repair Procedure

M. DeHarrera, Edwards Lifesciences LLC, Irvine, CA; W. Sun, University of Connecticut, Storrs, CT

Recently, a minimally invasive mitral valve repair procedure has been developed to mimic the surgical procedure of correcting mitral regurgitation (MR) with an annuloplasty ring. In this procedure, the device is crimped into a small diameter and percutaneously deployed via the patient's blood vessels into the coronary sinus (CS) and the great cardiac vein. Due to the location of the CS being parallel to the mitral annulus (MA), when the device contracts it shortens the MA and thus decreases MR.

In this study we simulate the deployment of the proximal anchor stent of the device into the CS. The CS is modeled as 1) displacement-controlled rigid wall, 2) porcine CS and 3) human CS. The porcine and human CS mechanical properties are obtained from vessel pressure-inflation experimental data of porcine and human CS tissues and modeled with hyperelastic models. The stent is modeled using a superelastic Nitinol material model. In addition, because of the importance of accurately calculating the peak strain on the stent surface, we also compare the following three ways of surface strain calculations: (1) integration points at the interior of the 3D brick elements, (2) averaged values at the nodal points extrapolated from the integration points and (3) the one integration point in each membrane surface element coating the free surface of the 3D elements.

We found that 1) a rigid surface driven model yields conservative results compared to a model embedded in a hyperelastic model with properties obtained from CS data, the former giving strain amplitudes that are significantly higher; 2) the mean and amplitudes of peak tensile strains will generally be higher in the coating membrane elements than in solid elements, regardless of whether the latter are reported at integration points or extrapolated to its nodal points. Such information will facilitate novel medical device designs.


Summary: In this study we simulate the deployment of the proximal anchor stent of a mitral valve repair device into the Coronary Sinus (CS). We compare various ways of modeling CS using 1) displacement-controlled rigid wall, 2) experimental porcine CS and 3) experimental human CS data. The stent is modeled using a superelastic Nitinol material model. With the simulation results, we quantify cyclic fatigue behaviors of the stent using Goodman methods and find that a rigid surface driven model yields conservative results compared to a model embedded in a hyperelastic model with properties obtained from CS data. Such information will facilitate novel medical device designs.