3-D Simulation
of the St. Jude Medical Bileaflet Valve Opening Process: Fluid-Structure
Interaction Study and Experimental Validation
Alberto Redaelli, Helene Bothorel, Emiliano Votta, Monica Soncini,
Umberto Morbiducci, Costantino Del Gaudio, Antonio Balducci, Mauro
Grigioni
Department of Bioengineering, Politecnico
of Milan, Milan, Laboratory of Biomedical Engineering, Istituto
Superiore di Sanità, Rome, Italy |
Background and aim of the study: Simulation of
the opening and closure dynamics of a mechanical valve through a moving
deforming mesh algorithm presents a challenge because of the large rotations
of the leaflet and of the small gaps between the housing and the leaflets,
which make remeshing a critical issue. The present study offers a computational
approach to the simulation of valve leaflet motion during the opening
process, together with an experimental set-up for validation of the model.
Methods: A fully 3-D simulation of the 27 mm St. Jude
Medical Hemodynamic Plus mechanical valve was performed using the
computational code, Fluent. Interaction between the leaflet and
fluid was simulated through customized user subroutines which,
according to a weakly coupled approach, update the leaflet velocities
through subsequent time steps by means of an under-relaxation procedure.
A parallel, experimental test was defined to collect data for the
set-up of simulations and for |
validation purposes.
Results: The computed leaflet velocity and angular displacement
compared well with experimental data. The model captured the main
features of the opening process, and did so also from a quantitative
viewpoint. Nonetheless, some discrepancies were observed, including
a delay of ~7 ms in the computed leaflet displacement and an underestimation
by ~7% of the maximum computed leaflet velocity.
Conclusion: The weakly coupled approach adopted here limited
computational costs, thus allowing the simulation of a fully 3-D
realistic mechanical valve within 154 CPU hours at minimal computational
costs. No significant drawbacks were raised in comparison with
the fully coupled approach. The opening process delay was similar
to that reported previously, and cannot be ascribed to the weakly
coupled approach adopted here.
The Journal of Heart Valve Disease 2004;13:804-813 |