A narrative review of computational approaches to understand type B aortic dissection mechanisms and predictions

Objective: This narrative literature review explores the use of computational fluid dynamics (CFD) methods to study type B aortic dissection (TBAD). This review aims to provide an overview of the principles, methodologies, and parameters used in these models, the current challenges and limitations in recent studies, and potential future directions within this field. Methods: Search terms included “aortic dissection,” “type B,” “computational fluid dynamics,” “fluid-structure interaction (FSI)” and “4D [four-dimensional] flow MRI [magnetic resonance imaging].” Additional articles were obtained from manual searches of the references found in the retrieved literature. Sources were selected based on relevance to the topic, including peer-reviewed articles, case reports, conference papers and preprints, ranging from older, foundational articles to more recent, forefront works through July 5, 2025. Results: For risk stratification and subsequent treatment, aortic dissection patients undergo imaging studies, such as computed tomography angiography, MRI, and 4D flow MRI. Typically, uncomplicated cases undergo dynamic studies, whereas urgent, complicated cases undergo static studies. These imaging data can be used to inform the morphology and boundary conditions of CFD models, creating patient-specific simulations to study TBAD and analyze wall shear stress, velocity profiles, pressure gradients, and flow patterns. In the CFD studies reviewed, the majority used patient-specific models; however, some implemented ideal, generic aortic models to analyze the hemodynamic changes associated with TBAD, examining aneurysmal degeneration, dissection propagation, effects of thrombosis, false lumen motion, and fenestrations. Other CFD studies examined hemodynamic changes after intervention (eg, thoracic endovascular aortic repair). Additional articles explored considerations for patient-specific CFD boundary conditions, tissue material-property assumptions, and CFD validation with in vitro phantom models. Small sample sizes were a common limitation across the studies. A small subset of studies used FSI with patient-specific models constructed from multiple imaging modalities for comprehensive TBAD analysis. Conclusions: Simulations integrating multiple imaging modalities and FSI with CFD have yet to be implemented consistently with large sample sizes. CFD and related predictive approaches can be used to analyze patient-specific data-integrated TBAD models to understand the short-term and long-term outcomes for individuals. The results from validated models can inform clinical decision-making for TBAD treatment, optimally stratifying treatment plan(s) to improve patient outcomes.