In vitro assessment of heart valve bioprostheses by cardiovascular magnetic resonance: four-dimensional mapping of flow patterns and orifice area planimetry.

OBJECTIVE: The hemodynamics in proximity to stented aortic bioprostheses still differ from that under physiological conditions. This may prevent desired cardiac remodeling and promote aortic diseases. Further improvements in prosthetic technology require an accurate survey of the flow conditions on the prosthetic level and in the ascending aorta. Cardiovascular magnetic resonance (CMR) may have the potential to provide more information by determining the prosthetic orifice area and visualizing the intravascular flow dynamics. We tested the feasibility to better characterize the hemodynamics of various stented bioprostheses in a pulsatile flow phantom by using CMR.

METHODS: The custom-made model consisting of a commercially available pump generating pulsatile flow, a tube system filled with a glycerin-water mixture, and a handcrafted bulbar-shaped cylinder holding the bioprostheses and simulating the aortic root, was located in a clinical 1.5T CMR system. In this study, 10 stented aortic bioprostheses were investigated (Perimount® 21, 23; Mitroflow® 19, 25; Hancock® 21, 23, 25; Mosaic® 21; Epic Supra® 21, 23). The prosthetic orifice area was visualized using steady-state free-precession cine imaging (spatial/temporal resolution 1.3×1.3×5 mm³/29 ms), quantified by manual planimetry and compared with published transthoracic echocardiographic data. Time-resolved three-dimensional phase-contrast flow mapping (1.8×1.8×3 mm³/45 ms) was applied to analyze the transprosthetic flow pattern.

RESULTS: Visualization of the prosthetic orifice area and the transprosthetic flow pattern was feasible in all prostheses. All orifice areas obtained by CMR in vitro were within one standard deviation of the mean of the published reference values obtained by echocardiography in vivo. Turbulent flow with vortex formation occurred both in proximity to the prosthesis and on the 'ascending aortic' level. Larger prosthetic sizes led to decreased flow velocities, but not mandatorily to less turbulences.

CONCLUSIONS: CMR allowed for a detailed interrogation of the fluid dynamics of various heart valve bioprostheses in a pulsatile flow model. It is an attractive tool to define proprietary reference values of the orifice area under standardized conditions and provides novel information regarding the flow pattern in proximity to the prosthesis.