Computational and Experimental Modeling of Subclinical Leaflet Thrombosis in Bioprosthetic Aortic Valves

PROJECT SUMMARY This project will devise experimentally and clinically validated computer models to elucidate the causal mecha- nisms of leaflet thrombosis in bioprosthetic heart valves (BHVs) following transcatheter or surgical aortic valve replacement, and thereby improve risk stratification and device selection. Each year, nearly 300,000 aortic valve replacements are performed worldwide to treat severe aortic valve stenosis, and the rate of valve replacement is projected to exceed 850,000/year by 2050. Traditionally, surgical aortic valve replacement (SAVR) was the gold standard for treating aortic valve stenosis; however, transcatheter aortic valve replacement (TAVR) has emerged as an alternative to SAVR that has been demonstrated to provide outcomes comparable to SAVR for elderly patients. Until recently, patients receiving aortic BHVs were thought to require limited anticoagulation, but in the past few years, clinical studies have unexpectedly revealed high rates of subclinical leaflet thrombosis (SLT) in BHVs after both SAVR and TAVR. SLT is associated with increased transient ischemic attacks and strokes, has been shown to trigger acute myocardial infarction, and is suspected to accelerate structural valve deterioration. Critically, SLT can progress to clinical valve thrombosis, which is a devastating complication. Wor- ryingly, a very recent study on two-year data for the PARTNER 3 trial found a statistically significant increase in valve thrombosis following TAVR compared to SAVR (2.6% post-TAVR vs. 0.7% post-SAVR, p=0.02). Two mechanisms have been hypothesized for the increased early incidence of SLT in TAVR: 1) abnormal blood flow patterns in the vicinity of the transcatheter aortic valve (TAV) (e.g., flow stasis, turbulence, paravalvular leak) and 2) stent-crimp induced injury of the TAV leaflets, which activates coagulation and platelet deposition. Although clinical imaging can detect SLT following aortic valve replacement, there is currently no approach to predict which patients will develop SLT following either SAVR or TAVR. The goal of this project is to devise patient-specific computational fluid-structure interaction (FSI) models of BHVs coupled to biochemically and biophysically detailed thrombosis models to characterize the mechanisms that lead to leaflet thrombosis and, ultimately, to predict leaflet thrombosis risk using clinical data in patients undergoing TAVR and SAVR. This project promises to transform computation-based methods for AVR device selection and SLT risk assessment. The project goals will be accomplished through three Specific Aims. Aim 1 focuses on experimental validation of FSI models; Aim 2 studies mechanisms that lead to leaflet thrombosis after aortic valve replacement; and Aim 3 focuses on clinical validation and device selection. Through these studies, a multidisciplinary team with an established record of collaboration will integrate mathematical, computational, experimental, and clinical ap- proaches to yield substantial innovation by establishing novel, rigorously validated models of flow, FSI, and thrombosis post-AVR that will ultimately enable patient-specific SLT risk assessment. Further, because throm- bosis are major challenges for many types of implanted devices, the project promises to have a broad impact.