Coaptation Surface Area and Aortic Regurgitation: The Infinite Potential of Finite Element Analysis

Abstract

MODELING OF HUMAN anatomic and physiologic processes is inherently challenging due to the complexity involved. Finite element (FE) modeling is a computer simulation technique for engineering and mathematical physics problems that breaks down a complex structure into a large number of smaller components (termed finite elements) tied together as an idealization of the true physical system. The simulation technique has been used in the evaluation of structural analysis, electromagnetic potential, and fluid flow, among others. Using the predictable behavior of individual FEs, the model then can be used to predict the behavior of the original structure. Changes then can be made to either the structure or the loading conditions to analyze behavior, which at least theoretically can predict outcome prior to intervention. FE analysis historically has found little ground in clinical practice, especially in the surgical realm. It is a time consuming process in which a certain amount of speculation is inherent. Moreover, simulation models of the heart do not, generally speaking, take into account that the surgeon has to conduct the procedure on a depressurized heart, using his or her intuition for a repair that will be tested in conditions different from when the procedure was performed. A trend has been observed in aortic valve repair that is based on visualization of the functional aortic annulus as a unit, encompassing the valve cusps, the ventriculo-aortic junction, the sinuses, and the sinotubular junction (STJ). Traditionally, one of the limiting factors in the widespread adoption of the repair techniques for the aortic valve has been the heterogeneity in both valve anatomy and reported surgical technique. As experience has grown, however, not only tricuspid, but even bicuspid aortic valves have been repaired successfully.1Svensson L.G. Al Kindi A.H. Vivacqua A. et al.Long-term durability of bicuspid aortic valve repair.Ann Thorac Surg. 2014; 97: 1539-1548Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar This progress also has sparked an interest in tools to predict postoperative valve performance using preoperative data. This is where FE analysis has come in, though much of it has been in the field of transcatheter aortic valve replacement (TAVR) and aortic stenosis. As work has been underway to test the long-term feasibility of TAVRs in aortic insufficiency (AI), by corollary FE analysis has been applied to study AI as well.2Ruparelia N. Prendergast B.D. TAVI in 2015: Who, where and how?.Heart. 2015; 101: 422-431Crossref Scopus (17) Google Scholar This has had implications in how repair can be guided by means other than pure surgical intuition. FE analysis also has shown abnormality in the stress experienced by seemingly normal aortic leaflets in patients with Marfan’s syndrome, which contributes to regurgitation as root dilation progresses.3Grande-Allen K.J. Cochran R.P. Reinhall P.G. et al.Mechanisms of aortic valve incompetence: Finite-element modeling of Marfan syndrome.J Thorac Cardiovasc Surg. 2001; 122: 946-954Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar In this issue, Sohmer et al have written about the FE analysis of human AV models of dilated aortic roots and have attempted to find a correlation between FE analysis–based calculation of the coaptation surface area (CoSA) and AI.4Sohmer B. Jafar R. Patel P. et al.Aortic valve cusp coaptation surface area by 3D transesophageal echocardiography correlates with severity of aortic valve insufficiency.J Cardiothorac Vasc Anesth. 2017; Abstract Full Text Full Text PDF Scopus (3) Google Scholar The authors also used data derived from 98 simulations of increasingly dilated aortic roots to validate the same correlation in 3D transesophageal echocardiography (TEE) data obtained from 10 patients. The authors report an increase in the severity of regurgitation that correlates with decreased CoSA with an increase in the dimensions of the functional aortic annulus and STJ. Similar results have been reported in experimental models in which dilation of the STJ and increased annular dimension resulted in reduction in the CoSA and an increase in effective regurgitant orifice area.5Toeg H. Jafar R. Ngu J. et al.The effect of ventriculo-aortic junction dilatation on aortic insufficiency.Can J Cardiol. 2017; 33: S89Abstract Full Text Full Text PDF Google Scholar With regard to FE analysis, one of the earliest uses of this technique to evaluate a cardiac valve was reported in 1993, which described how annular and papillary muscle contraction aids in mitral valve function.6Kunzelman K.S. Cochran R.P. Chuong C. et al.Finite element analysis of the mitral valve.J Heart Valve Dis. 1993; 2: 326-340PubMed Google Scholar The same group later demonstrated the increase in stress and delay in coaptation of the leaflets in cases of annular dilation.7Kunzelman K.S. Reimink M.S. Cochran R.P. Annular dilatation increases stress in the mitral valve and delays coaptation: A finite element computer model.Cardiovasc Surg. 1997; 5: 427-434Crossref PubMed Scopus (0) Google Scholar The FE modeling and its refinement using in vivo studies has been two-way traffic, with input from each improving the other. Admittedly, the mitral valve apparatus is a highly complex structure and presents a challenging re-creation. To increase the validity and accuracy of such methods, linking the computer-generated model to in vivo imaging such as computed tomography, TEE, and magnetic resonance imaging–based reconstruction has been carried out.8Gao H. Qi N. Feng L. et al.Modelling mitral valvular dynamics–current trend and future directions.Int J Numer Methods Biomed. 2017; Crossref PubMed Scopus (25) Google Scholar Thus, as imaging techniques improve, their contribution to creating robust FE models increases, and the reliability of these FE models improves. In this context, computer-based simulations represent a powerful tool that can provide predictive information aimed at outcome improvement from input at the conception (procedure feasibility) and the execution (intraoperative) stage. However, some ingredients require a degree of assumption for modelling of such complexity, either because they are almost impossible to measure (eg, mechanical tissue property) or because they are theoretical parameters.9Russ C, Hopf R, Hirsch S, et al. Simulation of transcatheter aortic valve implantation under consideration of leaflet calcification. 2013 35th Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. (EMBC), Osaka, 2013, pp. 711-714Google Scholar One of the constraints of previous work in FE technology is that in vivo performance of human tissue differs significantly from prosthetic material response under test conditions.10Votta E. Caiani E. Veronesi F. et al.Mitral valve finite-element modelling from ultrasound data: A pilot study for a new approach to understand mitral function and clinical scenarios.Philos Trans R Soc Lond Math Phys Eng Sci. 2008; 366: 3411-3434Crossref PubMed Scopus (98) Google Scholar This is an evolving science, and FE simulations trying to predict the success of TAVRs have tied in contact forces and medical performance of the device. The medical outcome of the device is dependent on the lateral friction between the stent and the surrounding tissue, which cannot be measured directly but can be estimated using simulation. It follows that simulated patient-specific contact forces can be used to identify the ideal size, type, and position of implant.11Hopf R. Gessat M. Russ C. et al.Finite element stent modeling for the postoperative analysis of transcatheter aortic valve implantation.J Med Devices. 2017; 11: 021002Crossref Scopus (7) Google Scholar This analogy then can be extended to the lateral stress that keeps the aortic cusps coapted during diastole. A higher area over which blood can exert lateral pressure during diastole will mean more force, which will keep the leaflets approximated and reduce regurgitation. When undertaking an analysis of this kind, one uncertainty remains: the specific material response from the tissue of individual patients, which in simulation is derived as a statistical average from the population.9Russ C, Hopf R, Hirsch S, et al. Simulation of transcatheter aortic valve implantation under consideration of leaflet calcification. 2013 35th Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. (EMBC), Osaka, 2013, pp. 711-714Google Scholar Questions have risen about the reproduction detail in the simulation with regard to native valve material, collagen, and tissue integrity and whether it replicates human tissue under real-life conditions. Sohmer et al have used a linear variability model for tissue elastance for the aortic cusps with porcine valvular data substituting for 18-year-old human tissue and, on the other end of the spectrum, 80-year-old human tissue. This should provide a reasonable template to devise surgical techniques in experimental models to optimize CoSA and avoid residual regurgitation. We now have proof of concept that will further spawn work to validate the normal descriptors in a larger cohort. What will, one hopes, eventually follow is the target values for CoSA to which the surgeon must strive to achieve optimal repair. An accurate, fast, and reliable AV apparatus construction based on real-time imaging is key to modeling that is aimed at altering clinical course, especially in the intraoperative time period. One can foresee a future in which CoSA calculations will be included in the pre- and postoperative assessment of the aortic valve where a repair is either envisioned or conducted. The strength of the current paper is that FE modeling and its results agreed with what was found with 3D TEE in real patients. Exporting 3D datasets and combining them with FE analysis to allow the surgeon to visualize the results and modify the plan prior to operation is obviously the holy grail of this process and is similar to work being done using 3D TEE with the mitral valve.12Ryan L.P. Jackson B.M. Eperjesi T.J. et al.A methodology for assessing human mitral leaflet curvature using real-time 3-dimensional echocardiography.J Thorac Cardiovasc Surg. 2008; 136: 726-734Abstract Full Text Full Text PDF PubMed Scopus (37) Google Scholar As identified by the authors, further work is needed to identify the predictive value of CoSA in avoiding long-term valve dysfunction after repair and how intervention guided by CoSA compares with traditional means.