Long-term prognostic impact of paravalvular leakage on coronary artery disease requires patient-specific quantification of hemodynamics

Figure 1. Electrical and anatomical schematic diagrams of the lumped parameter modeling. (a) Anatomical illustration; (b) Electrical depiction. This model includes the following sub-models: left main coronary artery, left anterior descending coronary artery, left circumflex coronary artery and right coronary artery, left ventricle, aortic valve, left atrium, mitral valve, aortic valve regurgitation, mitral valve regurgitation, systemic circulation, pulmonary circulation. Abbreviations are the same as in Table 2.

Figure 2. Schematic diagram of computational domain. Electrical and anatomical schematic diagrams of the patient-specific lumped parameter model that provides patient-specific boundary conditions for the fluid domain. This model includes the following sub-models. (1) ascending aorta, (2) left ventricle, (3) left anterior descending coronary artery, (4) left circumflex coronary artery, and (5) right coronary artery. Abbreviations are the same as in Table 2. Simulation domain and FSI modeling. Patient-specific LPM simulating the function of the left side of the heart and coronary arteries provided the patient-specific boundary conditions of the inlet and outlets.

Figure 3. Hemodynamics parameter: (a) Regurgitant flow velocity waveform at the paravalvular leakage cite (Apical 5 chamber view); (b) Parasternal short-axis color doppler view of the prosthetic valve and the paravalvular leakage area (Vena contracta area: 1.04 cm2) and its circumferential extent with respect to prosthetic valve diameter (35%); (c) Long axis color doppler view of paravalvular leakage jet interaction with diastolic flow behind the posterior mitral valve leaflet. Geometrical parameters: (d) Parasternal long-axis view associated with different parts of the aortic root and ascending aorta before TAVR; (e) Parasternal short-axis view of aortic valve before TAVR; (f) M-Mode measurement of ascending aorta before TAVR; (g) Parasternal long-axis view associated with different parts of the aortic root, prosthetic frame and ascending aorta after TAVR; (h) Parasternal short-axis view of TAVR; (i) M-Mode measurement of ascending aorta after TAVR.

Figure 4. Reconstructed 3-dimensional geometry in a patient with AS who received TAVR using computed tomography, schematic diagram of LPM pressure boundary conditions and FSI. (a) CT views (coronal, sagittal and axial) of the ascending aorta, coronary branches, sinus of Valsalva and aortic valve for pre-TAVR (left column) and post-TAVR (middle and right column) as well as the segmentation process and geometry reconstruction; (b) computational domain and boundary condition. FSI simulations performed during diastole. Patient-specific LPM simulating the coronary perfusion pressure (ascending aorta pressure—LV diastolic pressure). This data was obtained from patient-specific imaged based lumped parameter model (Fig. 1).

Figure 5. Patient-specific lumped parameter boundary conditions (BC) and strongly coupled FSI model flow chart.

Figure 6. Grid convergence analysis. (a) Pre-TAVR: velocity comparisons for different mesh resolutions at the peak of filling phase for different coronary cross sections (sections A,B and C): difference in average velocity between mesh#3 and mesh#4 is less than 0.3% and between mesh#4 and #5 is less than 0.2%—difference in maximum velocity between mesh#3 and mesh#4 is less than 0.5% and between mesh#4 and mesh#5 is less than 0.3%; (b) Post-TAVR: velocity comparisons for different mesh resolutions at the peak of filling phase for different coronary cross sections (sections A,B and C): difference in average velocity between mesh#3 and mesh#4 is less than 0.3% and between mesh#4 and #5 is less than 0.2%—difference in maximum velocity between mesh#3 and mesh#4 is less than 0.5% and between mesh#4 and mesh#5 is less than 0.3%; In the aortic root and coronary arteries, the blood flow is laminar and does not experience turbulence during the diastolic phase. In this study, for all 3 patients, we considered the blood flow to be laminar.

Figure 7. Grid convergence analysis. (a) TAWSS at the PVL location: TAWSS comparisons for different mesh resolutions at the peak of filling phase (maximum TAWSS): difference in peak TAWSS between mesh#3 and mesh#4 is 3% and between mesh#4 and #5 is 1.03%; (b) Von Mises stress: Von mises stress comparisons for different mesh resolutions of left coronary artery at the peak of filling phase: difference in maximum stress between mesh#3 and mesh#4 is 5.25% and between mesh#4 and #5 is 1.43%; (c) Plots for Maximum TAWSS and Maximum Von Mises stress versus the number of mesh elements.

Figure 8. Validation: Doppler-based LPM and FSI framework versus patients Doppler echocardiography data. (a) & (d) Regurgitant flow waveform during diastole in patients #1 and #2; (b) Parasternal short axis view of PVL during diastole in patient #1; (e) Parasternal long axis view of PVL during diastole in patient #2; (c) & (f) PVL flow velocity streamlines during diastole in patients #1 and #2.

Figure 9. Evolution of vortical structure following TAVR in the aortic root and neo-sinus regions in patient #1 during diastole. (a) Mid-planar velocity of left coronary cusp neo-sinus; (b) Mid-planar velocity of right coronary cusp neo-sinus; (c) Mid-planar velocity of non-coronary cusp neo-sinus ; (d) Blood stasis volume. Pre-TAVR: severe aortic stenosis (EOA = 0.6 cm2), coronary artery disease and hypertension, diastolic dysfunction, atrial fibrillation, ejection fraction: 29%, brachial pressures: 61 and 107 mmHg, forward LV stroke volume: 35 mL; Post-TAVR: aortic valve (EOA = 1.7 cm2), hypertension, moderate mitral regurgitation, diastolic dysfunction, atrial fibrillation, ejection fraction: 34%, brachial pressures: 86 and 130 mmHg, forward LV stroke volume: 62 mL.

Figure 10. Evolution of vortical structure following TAVR in the aortic root and neo-sinus regions in patient #2 during diastole. (a) Mid-planar velocity of left coronary cusp neo-sinus; (b) Mid-planar velocity of right coronary cusp neo-sinus; (c) Mid-planar velocity of non-coronary cusp neo-sinus; (d) Blood stasis volume. Pre-TAVR: severe aortic stenosis (EOA = 0.6 cm2), coronary artery disease and dyslipidemia, diastolic dysfunction, ejection fraction: 38%, brachial pressures: 54 and 107 mmHg, forward LV stroke volume: 74 mL; Post-TAVR: aortic valve (EOA = 1.8 cm2), diastolic dysfunction, ejection fraction: 51%, brachial pressures: 59 and 120 mmHg, forward LV stroke volume: 92 mL.

Figure 11. Time averaged wall shear stress of the aortic root during diastole for patients #1 and #2 in both pre and post intervention status.

Figure 12. 3D distribution contours of Mises stress at peak diastole in patient#1 and patient#2 between baseline and 90-day post-TAVR.