Computational Finite Element Analyses to Optimize Graft Sizing During Aortic Valve-Sparing Procedure

Background and aim of the study: Aortic valvesparing (AVS) procedures have been introduced to treat ascending aorta dilatation and aortic valve insufficiency in the presence of preserved native aortic valve leaflets. Although the surgical technique has been standardized, the choice of best type and size of Dacron graft to be used remains a matter of debate. Herein are presented preliminary results based on a patient-specific finite element model aimed at optimizing the Dacron prosthesis size and shape. Previously, finite element analysis (FEA) has been applied to investigate medical problems and, in particular, to better evaluate the pathophysiology of the aortic root. To date, however, such methodology has not been applied to the patient-specific evaluation of AVS postoperative results. Methods: The framework of the FEA study included four steps: (i) the creation of a mathematic model of the patient's aortic root; (ii) the creation of a model for two different Dacron grafts (the standard straight graft and a Valsalva graft), with sizes of each type ranging from 24 to 30 mm; (iii) a virtual computer-based simulation of the AVS procedure, using each graft; and (iv) a virtual computer-based simulation of the diastolic closure of the repaired valve and an evaluation of post-implant physiology, based on three parameters: the height of coaptation ratio (HCR); the length of coaptation ratio (LCR); and the distance between the central point of coaptation and the ideal geometrical centre (D-C). Results: The simulation results of post-implant performance of the aortic valve revealed that both HCR and LCR were decreased as the graft size was increased, but no significant differences were identified between two types of graft. In contrast, the Valsalva graft, when compared to the standard straight graft, led to a significant reduction in D-C. The results in terms of HCR, LCR and D-C recommended unequivocally, for the specific case under investigation, that a 30 mm straight graft and a 28 mm Valsalva graft would ensure the most physiological valve behavior for the patient under investigation. Conclusion: In evaluating the potential of a preoperative prediction of the optimal graft size, using FEA, the virtual simulation of the AVS procedure proved to be feasible and useful in predicting the postoperative physiology of the aortic root. In particular, this finite element model might have a clinical impact as may be used to optimize the surgeon's choice of prosthesis size.