Abstract
Thoracic aortic aneurysms (TAAs) arise from a combination of biological and mechanical factors. Current clinical guidelines use size and rate of expansion to stratify risk, but such metrics do not predict if an aneurysm will stabilize, grow, dissect, or rupture. Computational biomechanical models can provide insights into mechanisms of aneurysm behavior that would be difficult or impossible to capture in vivo. Here, we use a constrained mixture theory of growth and remodeling to simulate lesion progression while co-varying rate-dependent parameters that contribute to the natural history of aneurysm growth. This includes insults to the material structure and mechanosensitivity of the vessel. This framework successfully simulates clinically-relevant phenotypes, including cases where lesions with initially similar degrees of dilatation or rates of expansion diverge in behavior later in their progression. By capturing this spectrum of outcomes, our framework lays a foundation for more accurate, patient-specific risk prediction and future integration of machine learning tools to accelerate translation into clinical practice.