Hemodynamic testing using three-dimensional printing and computational fluid dynamics preoperatively may provide more information in mitral repair than traditional image dataset

Figures

• 

  Figure 1. Three-dimensional digital model of the left heart based on a multimodality image data set. (A,B,C,D) CTA DICOM images were postprocessed to obtain a digital left heart cavity. (E,F,G,H) Mitral valve model based on fusion of 3D-TEE images to view the whole left heart system. 3D-TEE, three-dimensional transesophageal echocardiography; CTA, computed tomography angiography; DICOM, digital imaging and communication in medicine.

• 

  Figure 2. Workflow of creating the flexible 3D-printed model. (A) Hollow mold of heart model; (B) hollow mold is 3D-printed using soluble material-PVA; (C,D) final flexible 3D model made of silicone. LA, left atrium; LV, left ventricle; LAA, left atrial appendage; Ao, aorta. *, lesion area.

• 

  Figure 3. MCS. (A) General overview of the MCS. Compressed gas was generated from air pump and regulated by PLC. The 3DP heart model in a pressure chamber contracts and relaxes with compressed gas moving in and out. A pressure-elevating water tower is used to simulate afterload in systole. (B) Cutaway view showing gas transfer including the airflow orifice (red arrow) and chordae adjuster (black arrow). PLC, programmable logic controller; MCS, mock circulatory system; 3DP, three-dimensional printing.

• 

  Figure 4. CFD simulation of mitral regurgitation. (A) Mesh of the left heart system. (B) Stress analysis of prolapsed P3 mitral valve. Maximal stress mainly distributed at the boundary of P2 and P3, as well as the mitral annulus adjacent to P3. (C) The flow stream of regurgitation in systole. Consistent with in vivo, there was a severe eccentric regurgitation stream flowing in the direction of the anterior mitral valve, reaching the top of the left atrium at 0.20 seconds. CFD, computational fluid dynamics.

• 

  Figure 5. Surgical simulation using CFD (A,B,C,D) and three-dimensional model (E,F,G,H). (A,E) Resection of lesion area; (B,F) repaired leaflet; (C,G) annuloplasty ring; (D,H) annuloplasty ring tied in place. CFD, computational fluid dynamics.

• 

  Figure 6. Bland-Altman plots of coherence before and after surgical simulations. The x-axis is the mean VCW assessed in vivo, and the y-axis is the difference in VCW between in vitro and in vivo. The dots are distributed evenly above and below the average line (blue line), which was very close to y =0 (brown line). Therefore, we assumed that the in vitro data were consistent with the in vivo data. VCW, vena contrata width.

• 

  Figure 7. Accuracy of in vitro hemodynamics testing. The grade of regurgitation and the VCW measured in vitro and in vivo had high consistency, both before and after surgery. In addition, stress analysis in CFD showed that the mean stress on the mitral valve had reduced significantly from 1.18 to 0.59 MPa after surgery. CFD, computational fluid dynamics; VCW, vena contrata width.

• 

  Figure 8. Surgical simulation of patient with Barlow disease, using different PML lengths from 14.0 to 19.0 mm (with a 0.5-mm increase on each attempt). The results show that SAM as well as LVOT obstruction occurred when the PML length reached 16.5 mm, with a residual length of AML at 4.2 mm. AML, anterior mitral leaflet; LVOT, left ventricular outlet; PML, posterior mitral leaflet; SAM, systolic anterior motion.

• 

  Figure 9. Surgical simulation of patient with small left heart cavity. Plan A was MVR without any intervention to the interventricular septum. Plan B was a 0.5-mm decrease in interventricular septal thickness (red arrow) before mitral repair. Plan C was a 1.0-mm decrease in interventricular septal thickness (black arrow) before mitral repair. The result showed that the best surgical outcome occurred when the septal thickness was less than 12.0 mm (Plan C).