Health-e-Child - IST-2004-027749 - Deliverable D.11.4

Heart Diseases

Personalised Simulation of Myocardium Electromechanics
and Pulmonary Valve Replacement Surgery in
Repaired Tetralogy of Fallot: a Case Study

4. Simulating Pulmonary Valve Replacement Therapies

4.1. Medical Context

In the previous sections, we have described how we personalise an electromechanical model of the heart to simulate patient cardiac function. We now have a virtual heart that represents the pathophysiology of the patient. We can use this tool to simulate therapies, in particular we can test the two pulmonary valve replacement (PVR) strategies presented in the introductory section.

Pulmonary valve replacement is becoming a key therapy for patients suffering from chronic pulmonary valve regurgitations, and in particular for patients suffering from repaired Tetralogy of Fallot. The therapy consists in replacing new valves between the right-ventricle outflow tract and the pulmonary artery to remove the regurgitations. Nowadays, two competing approaches are available:

1. Valve replacement alone. Only the valves are replaced. If the right ventricle is too dilated, the surgeon may have to reduce the diameter of the right-ventricle outflow tract to place the valves, but no extra surgery is performed. The heart adapts itself to its new condition by natural remodelling.
2. Valve replacement with direct right-ventricle remodelling. The surgeon replaces the valves and he directly resects any myocardium regions that are impaired by the pathology. Scars and fibrosis are removed to improve the efficiency of the cardiac muscle. Right ventricle is explicitly reduced (del Nido et al., 2006).

However, the effects of these therapies on the cardiac function of the patient cannot be predicted easily. No clinical evidences are available to help the cardiologist in choosing the appropriate therapy. Therefore, we investigate in Health-e-Child how the virtual heart can be used to simulate the PVR therapies and to predict their effects on the cardiac function. The therapies are simulated as follows:

1. Valve replacement. Because replacing the valves only affects the regurgitations, we simulate this intervention by only disabling the regurgitations in the models.
2. Direct remodelling of the right-ventricle. Virtual soft-tissue intervention is carried out using SOFA framework as explained below.

4.2. Virtual Surgery

In our example, virtual surgery consists in removing the dyskinetic area we have observed in the images. As suggested by the biomechanical parameters we found when personalising the model, this area presents with a very low contractility, probably due to fibrosis or to the aneurysm. By removing this impaired area we hope to improve the efficiency of the myocardium.

More precisely, the virtual surgery is made up of three interactive and real-time steps:

1. Resecting the impaired myocardium tissue: we remove the aneurysm and the surrounding tissue that might be affected by fibrosis. This task is carried out through a point-and-click approach. The user first points at the center of the area to resect, then at the boundary, and all the mesh tetrahedra that lay within a sphere defined by the two points are removed (see movie below).
2. Remodelling the right ventricle: the free-wall is smoothly and carefully brought near the septum, to close the space resulting from the resections. The user clicks to grab the muscle and moves it. Simulation time-step can be modified in real-time to enable smooth and controlled displacements. Indeed, this operation is very sensitive since the virtual heart is empty of blood, which makes the remodelling complex. Furthermore, the final shape of the organ must be anatomically plausible despite of the resections.
3. Suturing the myocardium tissue: we sew the myocardium to definitively close the right-ventricle. After suture, the scar is defined to simulate the possible effects of the surgical intervention in that region, where electrical conductivity and myocardium contractility may result damaged by the intervention.

Virtual RV volume reduction surgery. From left to right: original mesh, after resection, during attachment, final mesh. Colour lines: fibre orientations. Black area: postoperative scar.

4.3. Results

Once the intervention is performed on the anatomical model, we run the electromechanical simulation using the same parameters as those estimated during the personalisation, except for the regurgitations, which are disabled (to simulate the replacement of the valves).

Volumes and pressures are computed at each time step. From the results, we can conclude that, for this patient, the best therapy would be replacing the valves and directly remodelling the myocardium. With this strategy indeed, the right ventricle volume is reduced, as a direct consequence of the intervention, but the right ejection fraction is significantly higher. Furthermore, left ventricle function is also improved, which highlights the relationship between the two ventricles. Finally, the right ventricle pressure at end-systole slightly decreases because of the scar, which modifies the myocardium motion in this area and decreases its efficiency.

RV volume curves and pressure-volume loops: (solid green) from segmentation; (solid black) preoperative simulation; (dashed blue) simulation after PPVR; (dash-dotted red) simulation after PVR with volume reduction. Vertical bars delineate the cardiac phases. Regurgitations are visible during the isovolumetric phases, when volume should be constant. As expected, RV volume reduction resulted in a shift of the curves.

4.4. Conclusions

For this patient, we found that valve replacement with volume reduction would provide a better immediate postoperative outcome. Probable reasons may be the removal of the dyskinetic area and the direct reduction of the RV volume. However, this procedure is very invasive and may be hazardous for the patient, with possible postoperative side-effects such as electro-physiological troubles due to the surgical scar. On the other hand, the effects of valve replacement alone are often visible late after replacement, after the myocardium adapts itself to its new loading conditions. A model of the natural remodelling of the myocardium is thus important to simulate the long-term postoperative effects of valve replacement.

References

• del Nido, P. J.: Surgical Management of Right Ventricular Dysfunction Late After Repair of Tetralogy of Fallot: Right Ventricular Remodeling Surgery, Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu. 9(1), 29-34, 2006