Risk prediction for thoracic aortic dissection: Is it time to go with the flow?

Central MessageAccurate risk prediction in thoracic aortic dissection must involve diagnostics congruent with aortic physiology. Aortic flow dynamics may carry the most promise to help model aortic wall behavior. Accurate risk prediction in thoracic aortic dissection must involve diagnostics congruent with aortic physiology. Aortic flow dynamics may carry the most promise to help model aortic wall behavior. The identification of individuals at risk for acute aortic disease, by virtue of its silent pathogenesis, is vital for patient outcome. However, ineffective screening methods hamper the efforts of patient identification, mostly due to the lower prevalence of thoracic aortic disease (relative to other cardiovascular conditions) and the lack of widely available clinical methods to assess physiological function—the aorta does not lend itself easily to functional assessment. Unlike the heart (for which motion is concordant with disease severity), which allows for accurate measurements of stress/strain, thickness, and all the numerical derivatives of these functions, the expansion and recoil of the aortic wall is small, and measurements of wall thickness are inaccurate with current imaging modalities. The primary mechanism of thoracic aortic dissection is mechanical material failure. More specifically, acute aortic dissection occurs where the stress created by the hemodynamic flow supersedes the strength of the aortic wall at an instantaneous time point.1Humphrey J.D. Schwartz M.A. Tellides G. Milewicz D.M. Role of mechanotransduction in vascular biology: focus on thoracic aortic aneurysms and dissections.Circ Res. 2015; 116: 1448-1461Crossref PubMed Scopus (268) Google Scholar Moving forward, the linkage of thoracic aortic flow with wall mechanical properties will be the key to understanding both aneurysmal growth and acute wall failure (ie, the initiation of dissection/rupture) and is certain to influence future guidelines for intervention. Research into aortic flow dynamics and wall mechanics is gaining momentum; however, it is yet to penetrate recommended guidelines or routine clinical practice. Much of this is due to the need for resource-heavy computational methods, or the lack of patient-specific data, or a combination of both. This paper summarizes the challenges to the assessment of thoracic aortic aneurysms (TAAs), the prediction of type A aortic dissection (TAAD), the utility of flow dynamics, and the likelihood of its role in clinical practice. The negative aortic wall remodeling that occurs during aneurysm progression does not limit the aorta's vital role in hemodynamic conduction and capacitance. Unlike chronic disease states of other organs that give rise to characteristic signs and symptoms (of deteriorating function), the chronic degenerative process in the media of the expanding aortic wall is indeed silent and only physically manifests in the patient at an acute point of wall mechanical failure (aortic dissection). Consequentially, there are no associated symptoms with aortic disease up until the acute event. This emphasizes the importance of more-advanced predictive models to manage this patient group. However, as a risk-prediction tool, size criteria have been shown to underestimate the risk of TAAD: more than 50% of dissections have been found to occur below recommended thresholds.2Pape L.A. Awais M. Woznicki E.M. Suzuki T. Trimarchi S. Evangelista A. et al.Presentation, diagnosis, and outcomes of acute aortic dissection: 17-year trends from the international registry of acute aortic dissection.J Am Coll Cardiol. 2015; 66: 350-358Crossref PubMed Scopus (707) Google Scholar,3Rylski B. Blanke P. Beyersdorf F. Desai N.D. Milewski R.K. Siepe M. et al.How does the ascending aorta geometry change when it dissects?.J Am Coll Cardiol. 2014; 63: 1311-1319Crossref PubMed Scopus (164) Google Scholar The more we learn about thoracic aortic biomechanics, the more we realize that aortic diameter is grossly inadequate as a predictor of prospective aortic dissection or rupture. Medial degeneration has been found to occur independently of aortic diameter.4Salmasi M.Y. Pirola S. Sasidharan S. Fisichella S.M. Redaelli A. Jarral O.A. et al.High wall shear stress can predict wall degradation in ascending aortic aneurysms: an integrated biomechanics study.Front Bioeng Biotechnol. 2021; 9: 750656Crossref PubMed Scopus (22) Google Scholar Increasing aortic size is only one of the manifestations of aortopathy: aneurysm diameter is a consequence of negative aortic remodeling but not its cause. Although there is truth that overall aortic wall tension increases with diameter (vis-à-vis: law of Laplace), it does not account for heterogenous wall thickness and stress distribution throughout the aortic segment of interest.4Salmasi M.Y. Pirola S. Sasidharan S. Fisichella S.M. Redaelli A. Jarral O.A. et al.High wall shear stress can predict wall degradation in ascending aortic aneurysms: an integrated biomechanics study.Front Bioeng Biotechnol. 2021; 9: 750656Crossref PubMed Scopus (22) Google Scholar As well as the absent warning signs/symptoms arising from TAA progression, the overreliance on size criteria represents a gross unmet clinical need. Of course, tremendous advancements in the genetic basis of thoracic aortopathy have been made in the last 3 decades, including both syndromic and nonsyndromic causes of aortopathy. There have been several implicated mutations, many of which have, correctly so, directly affected guidelines (by relaxing the size criteria for affected patients).5Brownstein A. Kostiuk V. Ziganshin B. Zafar M. Kuivaniemi H. Body S. et al.Genes associated with thoracic aortic aneurysm and dissection: 2018 update and clinical implications.Aorta (Stamford). 2018; 6: 13-20Crossref PubMed Scopus (64) Google Scholar However, the main limitation of routine genetic testing for TAA and TAAD risk-assessment is the low pick-up rate: the majority of subjects with TAA (>80%) have sporadic forms of the disease with no identifiable genetic mutations,6Ziganshin B.A. Bailey A.E. Coons C. Dykas D. Charilaou P. Tanriverdi L.H. et al.Routine genetic testing for thoracic aortic aneurysm and dissection in a clinical setting.Ann Thorac Surg. 2015; 100: 1604-1611Abstract Full Text Full Text PDF PubMed Scopus (114) Google Scholar again strengthening the impetus for research into functional biomarkers. The application of fluid mechanics to cardiovascular physiology has been made possible through the emergence of 2 main technologies: (1) 4-dimensional flow-sensitive magnetic resonance imaging (4D-flow MRI); and (2) computational fluid dynamics (CFD).7Jarral O.A. Tan M.K.H. Salmasi M.Y. Pirola S. Pepper J.R. O’Regan D.P. et al.Phase-contrast magnetic resonance imaging and computational fluid dynamics assessment of thoracic aorta blood flow: a literature review.Eur J Cardiothorac Surg. 2019; 57: 438-446Google Scholar CFD has already made strides in cardiovascular diagnostics, via its commercialized use for noninvasive coronary flow estimation using computed tomography–derived measurements of fractional flow reserve.8Lee J.M. Choi G. Koo B.K. Hwang D. Park J. Zhang J. et al.Identification of high-risk plaques destined to cause acute coronary syndrome using coronary computed tomographic angiography and computational fluid dynamics.JACC Cardiovasc Imaging. 2019; 12: 1032-1043Crossref PubMed Scopus (185) Google Scholar Although numerous aortic flow patterns can be analyzed, the literature focuses mainly on wall shear stress (WSS): the tangential force (per unit area) exerted by flowing blood on the surface of the vessel (in this case the aorta). WSS has an important role in aortic homeostasis due to its direct impact on the aortic endothelium. Recent discoveries have allowed us to attribute aneurysm pathogenesis to the disruption in mechanical homeostasis and maladaptive alterations in the composition of structural proteins within the extracellular matrix, in direct response to altered blood flow.1Humphrey J.D. Schwartz M.A. Tellides G. Milewicz D.M. Role of mechanotransduction in vascular biology: focus on thoracic aortic aneurysms and dissections.Circ Res. 2015; 116: 1448-1461Crossref PubMed Scopus (268) Google Scholar This process of mechanotransduction underpins our understanding of aortic wall homeostasis, as well as disease pathogenesis. Exact pathways are still being discovered, with the knowledge of genetic malformations directly informing our understanding of integrins, intracellular proteins, and structural proteins. In this light, analysis of flow can provide vital information on the downstream mechanobiological response and mechanical integrity of the aortic wall. The difficulty, however, is to accurately match flow with aortic wall response and stress/strain (as will be explained later). WSS can either be measured directly from 4D-flow MRI or using CFD simulations. Studies have shown the relative underestimation of numerical WSS from the direct method, although both CFD and 4D-flow MRI techniques generate similar WSS distribution maps.9Piatti F. Sturla F. Bissell M.M. Pirola S. Lombardi M. Nesteruk I. et al.4D flow analysis of BAV-related fluid-dynamic alterations: evidences of wall shear stress alterations in absence of clinically-relevant aortic anatomical remodeling.Front Physiol. 2017; 8: 441Crossref PubMed Scopus (52) Google Scholar The largest study carried out assessing WSS in healthy ascending aortas was published by Callaghan and Grieve in 2018,10Callaghan F.M. Grieve S.M. Translational physiology: normal patterns of thoracic aortic wall shear stress measured using four-dimensional flow MRI in a large population.Am J Physiol Heart Circ Physiol. 2018; 315: H1174-H1181Crossref PubMed Scopus (33) Google Scholar which reported a number of key findings from 224 normal aortas using 4D-flow MRI (Figure 1). Most importantly, the spatial distribution of WSS was highly heterogeneous, with a localized region of elevated WSS along the length of the anterior wall seen across all individuals. Average peak systolic WSS was 1.79 ± 0.71 Pa in the aortic arch and was significantly greater at 2.23 ± 1.04 Pa in the descending aorta, having a strong negative correlation with advancing age. A number of important findings have been made concerning WSS quantification in TAA disease (Table 1):1.Lower peak WSS—compared with healthy volunteers, TAA exhibited lower WSS at peak systole size,11Bieging E.T. Frydrychowicz A. Wentland A. Landgraf B.R. Johnson K.M. Wieben O. et al.In vivo three-dimensional MR wall shear stress estimation in ascending aortic dilatation.J Magn Reson Imaging. 2011; 33: 589-597Crossref PubMed Scopus (93) Google Scholar,14Van Ooij P. Potters W.V. Nederveen A.J. Allen B.D. Collins J. Carr J. et al.A methodology to detect abnormal relative wall shear stress on the full surface of the thoracic aorta using four-dimensional flow MRI.Magn Reson Med. 2015; 73: 1216-1227Crossref PubMed Scopus (64) Google Scholar which was inversely related to aortic size.2.Greater time-averaged WSS—time-averaged WSS was often greater than in controls due to a decreased systolic to diastolic WSS ratio, and delayed onset of peak WSS.11Bieging E.T. Frydrychowicz A. Wentland A. Landgraf B.R. Johnson K.M. Wieben O. et al.In vivo three-dimensional MR wall shear stress estimation in ascending aortic dilatation.J Magn Reson Imaging. 2011; 33: 589-597Crossref PubMed Scopus (93) Google Scholar3.Asymmetric WSS distribution—maximal peak WSS was often observed on the anterior wall of the proximal aorta and lowest on the proximal inner curve, and the right outer curve just before the arch.11Bieging E.T. Frydrychowicz A. Wentland A. Landgraf B.R. Johnson K.M. Wieben O. et al.In vivo three-dimensional MR wall shear stress estimation in ascending aortic dilatation.J Magn Reson Imaging. 2011; 33: 589-597Crossref PubMed Scopus (93) Google Scholar,14Van Ooij P. Potters W.V. Nederveen A.J. Allen B.D. Collins J. Carr J. et al.A methodology to detect abnormal relative wall shear stress on the full surface of the thoracic aorta using four-dimensional flow MRI.Magn Reson Med. 2015; 73: 1216-1227Crossref PubMed Scopus (64) Google Scholar4.Eccentric WSS patterns—this was particularly observed in patients with Marfan syndrome, even in the absence of aneurysmal or valve disease. Patients with Marfan syndrome were also observed to have the greatest eccentricity of WSS patterns at the inner curve of the proximal aorta, and more anteriorly in its distal part.12Geiger J. Markl M. Herzer L. Hirtler D. Loeffelbein F. Stiller B. et al.Aortic flow patterns in patients with Marfan syndrome assessed by flow-sensitive four-dimensional MRI.J Magn Reson Imaging. 2012; 35: 594-600Crossref PubMed Scopus (67) Google Scholar,15Wang H.H. Chiu H.H. Tseng W.Y.I. Peng H.H. Does altered aortic flow in marfan syndrome relate to aortic root dilatation?.J Magn Reson Imaging. 2016; 44: 500-508Crossref PubMed Scopus (23) Google Scholar Greater oscillatory shear index was observed at the sinotubular junction and in the arch.12Geiger J. Markl M. Herzer L. Hirtler D. Loeffelbein F. Stiller B. et al.Aortic flow patterns in patients with Marfan syndrome assessed by flow-sensitive four-dimensional MRI.J Magn Reson Imaging. 2012; 35: 594-600Crossref PubMed Scopus (67) Google Scholar,15Wang H.H. Chiu H.H. Tseng W.Y.I. Peng H.H. Does altered aortic flow in marfan syndrome relate to aortic root dilatation?.J Magn Reson Imaging. 2016; 44: 500-508Crossref PubMed Scopus (23) Google ScholarTable 1Summary of studies modeling flow in the aorta in health and diseaseStudyNumber of patientsAims of studyTechnique used, variables calculated, software usedKey findings/useful imagesStudies analyzing ascending aortic flow Bieging et al, 201111Bieging E.T. Frydrychowicz A. Wentland A. Landgraf B.R. Johnson K.M. Wieben O. et al.In vivo three-dimensional MR wall shear stress estimation in ascending aortic dilatation.J Magn Reson Imaging. 2011; 33: 589-597Crossref PubMed Scopus (93) Google Scholar (423)21Assessment of impact of aortic dilatation on flow patterns.4D-flow MRIWSS, streamlines, pathlines, helical and vertical flowMATLAB (in-house; MathWorks)EnSight (ANSYS, Inc)Compared with controls, patients with ascending aorta dilatation had:•greater time averaged WSS, and diastolic WSS;•decreased systolic to diastolic WSS ratio;•delayed onset of peak WSS; and•decreased peak systolic WSS.Maximal WSS was found to be on the anterior wall of the ascending aorta. Increased vertical and helical flow was also found during diastole. Geiger et al, 201212Geiger J. Markl M. Herzer L. Hirtler D. Loeffelbein F. Stiller B. et al.Aortic flow patterns in patients with Marfan syndrome assessed by flow-sensitive four-dimensional MRI.J Magn Reson Imaging. 2012; 35: 594-600Crossref PubMed Scopus (67) Google Scholar (161)34Comparison of patients with Marfan syndrome and healthy volunteers.4D-flow MRIParticle traces, streamlines, helical and vertical flowEnSight (ANSYS, Inc)Local helix flow in the ascending aorta was significantly more common in Marfan syndrome (without aneurysm or valve insufficiency) in comparison with healthy volunteers (with similar sized aortas).Helix and vertical flow in the descending aorta were more common in patients with Marfan syndrome. Barker et al, 201813Barker A.J. Markl M. Fedak P.W.M. Assessing wall stresses in bicuspid aortic valve-associated aortopathy: forecasting the perfect storm?.J Thorac Cardiovasc Surg. 2018; 156: 471-472Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar (471)42Comparison of flow patterns between patients with aortic dilatation and healthy controls.4D-flow MRILaminar viscous energy lossVMTK (Orobix)ImageJ (National Institutes of Health)Viscous energy loss was significantly greater in patients with dilated thoracic aortas and in those with aortic stenosis, when compared with healthy volunteers. Van Ooij et al, 201514Van Ooij P. Potters W.V. Nederveen A.J. Allen B.D. Collins J. Carr J. et al.A methodology to detect abnormal relative wall shear stress on the full surface of the thoracic aorta using four-dimensional flow MRI.Magn Reson Med. 2015; 73: 1216-1227Crossref PubMed Scopus (64) Google Scholar (425)30Cohort-averaged systolic WSS calculated for a group of volunteers, aneurysm and aortic stenosis patients.4D-flow MRIWSS, cohort-averaged systolic WSSMATLAB (in-house; MathWorks),Mimics (Materialise), Linear Image Registration Tool (FLIRT; FMRIB)Patients with aortic dilatation showed significantly lower WSS on 7% of the ascending aorta (distal outer curvature and proximal inner curvature), and stenosis patients showed significantly greater WSS on 34% of the ascending aorta (entire outer curvature). Wang et al, 201615Wang H.H. Chiu H.H. Tseng W.Y.I. Peng H.H. Does altered aortic flow in marfan syndrome relate to aortic root dilatation?.J Magn Reson Imaging. 2016; 44: 500-508Crossref PubMed Scopus (23) Google Scholar (426)32Comparison of patients with Marfan syndrome and healthy volunteers.4D-flow MRIWSS, OSI, nonroundness (measure of asymmetry of segmental WSS), vorticity, helicityMATLAB (in-house; MathWorks)EnSight (ANSYS, Inc)Patients with Marfan syndrome had lower axial WSS in the root and lower circumferential WSS in the arch compared with healthy patients. They also demonstrated greater OSI (axial and circumferential) at the sinotubular junction and arch, but lower OSI (circumferential) in the descending aorta.WSS was strongly correlated with body surface area and aortic size in patients with Marfan syndrome. Callaghan and Grieve, 201810Callaghan F.M. Grieve S.M. Translational physiology: normal patterns of thoracic aortic wall shear stress measured using four-dimensional flow MRI in a large population.Am J Physiol Heart Circ Physiol. 2018; 315: H1174-H1181Crossref PubMed Scopus (33) Google Scholar224Analysis of normal patterns of wall shear stress in a large population of healthy volunteers4D-flow MRIWSS, OSISpatial distribution of WSS was heterogenous: A localized region of elevated WSS along the length of the anterior wall seen across all individualsAverage peak systolic WSS was 1.79 ± 0.71 Pa in the aortic arch and significantly greater (2.23 ± 1.04 Pa) in the descending aorta, with a strong negative correlation with advancing age.Studies combining aortic flow analysis with aortic wall material properties Trabelsi et al, 201516Trabelsi O. Davis F.M. Rodriguez-Matas J.F. Duprey A. Avril S. Patient specific stress and rupture analysis of ascending thoracic aneurysms.J Biomech. 2015; 48: 1836-1843Crossref PubMed Scopus (53) Google Scholar5To analyze the deformation and stress distributions in patient-specific ATAA modelsCalculation of retrospective rupture risk index: peak wall stress divided by peak systolic blood pressureECG-gated dynamic CT scansPatient-specific geometry extraction (Materialise Mimics)Excised aortic samples subjected to bulge inflationFinite element simulations (Abaqus; Dassault Systèmes)Peak wall stress values for 5 patients were 783, 652, 438, 601, and 4122 kPa.Peak wall stresses were only loosely correlated with maximum aneurysm diameter for each patient.Very large aneurysms also had lower peak wall stresses. Trabelsi et al, 201817Trabelsi O. Gutierrez M. Farzaneh S. Duprey A. Avril S. A non-invasive methodology for ATAA rupture risk estimation.J Biomech. 2018; 66: 119-126Crossref PubMed Scopus (15) Google Scholar13Distensibility, membrane tangential stiffnessECG-gated dynamic CT scansPatient-specific geometry extraction (Materialise Mimics)Excised aortic samples subjected to bulge inflationFinite element simulations (Abaqus; Dassault Systèmes)Inverse relationship between aortic distensibility and age/aneurysm diameter. Bollache et al, 201818Bollache E. Guzzardi D.G. Sattari S. Olsen K.E. Di Martino E.S. Malaisrie S.C. et al.Aortic valve-mediated wall shear stress is heterogeneous and predicts regional aortic elastic fiber thinning in bicuspid aortic valve-associated aortopathy.J Thorac Cardiovasc Surg. 2018; 156: 2112-2120Abstract Full Text Full Text PDF PubMed Scopus (94) Google Scholar27Patients with BAV aortopathy4D-flow MRIWSS quantification from 4D flowExplanted aortic tissue subjected to regional mechanical testingQuantitative histopathology and mechanical tensile testingElastic fibers were thinner and WSS was greater along the greater curvature compared with other circumferential regions.Increased regional WSS was associated with decreased elastic fiber thickness. Patient stratification with subanalysis showed an increase in the correlation between WSS and histopathology with aortic valve stenosis and smaller aortic diameters. Elastic fiber thinning was associated with circumferential stiffness Campobasso et al, 201819Campobasso R. Condemi F. Viallon M. Croisille P. Campisi S. Avril S. Evaluation of peak wall stress in an ascending thoracic aortic aneurysm using FSI Simulations: effects of aortic stiffness and peripheral resistance.Cardiovasc Eng Technol. 2018; 9: 707-722Crossref PubMed Scopus (51) Google Scholar1Distributions of blood flow, WSS, and wall stress were evaluated in the ascending thoracic aorta using the FSI analyses.FSI modelsPatient-specific geometries and boundary conditions derived from 4D-flow MRISignificant flow eccentricity in the simulations, in very good agreement with velocity profiles measured using 4D-flow MRI.Significant increase of peak wall stress due to the increase of peripheral resistance and aortic stiffness.The largest peripheral resistance (1010 kg.s.m−4) and stiffness (10 MPa) resulted in a maximal principal stress equal to 702 kPa, whereas it was only 77 kPa in normal conditions Condemi et al, 202020Condemi F. Campisi S. Viallon M. Croisille P. Avril S. Relationship between ascending thoracic aortic aneurysms hemodynamics and biomechanical properties.IEEE Trans Biomed Eng. 2020; 67: 949-956Crossref PubMed Scopus (18) Google Scholar10Comparison between CFD-calculated vs 4D-flow MRI-measured velocities and flowsAnalyze the relationship between WSS parameters and tensile parameters of the aortic wall4D-flow MRI and CT angiograms to provide patient-specific segmented aortic geometries (CRIMSON) and generate CFD simulations (EnSight; ANSYS)Explanted aneurysms subjected to bulge inflation testsThere was a significant positive correlation between time averaged WSS and rupture stretchThere was a significant negative correlation between elastic modulus and WSS. Jamaleddin Mousavi et al, 202121Jamaleddin Mousavi S. Jayendiran R. Farzaneh S. Campisi S. Viallon M. Croisille P. et al.Coupling hemodynamics with mechanobiology in patient-specific computational models of ascending thoracic aortic aneurysms.Comput Methods Programs Biomed. 2021; 205: 106107Crossref PubMed Scopus (17) Google Scholar4Couple patient-specific hemodynamics of ATAAs, with a computational G&R modelComparison of 2 healthy and 2 subjects with disease4D-flow MRIPatient-specific CFD simulations (MATLAB [MathWorks], EnSight; ANSYS).Finite element model of growth and remodeling (Abaqus; Dassault Systèmes)—using homogenous wall thicknessThe evolution of wall stiffness was shown to be a major risk factor for ATAAs.G&R parameters, such as the rate of collagen production or cell mechanosensitivity, play a critical role in ATAA progression and remodeling. Salmasi et al, 20214Salmasi M.Y. Pirola S. Sasidharan S. Fisichella S.M. Redaelli A. Jarral O.A. et al.High wall shear stress can predict wall degradation in ascending aortic aneurysms: an integrated biomechanics study.Front Bioeng Biotechnol. 2021; 9: 750656Crossref PubMed Scopus (22) Google Scholar10Relationship between regional WSS and material properties were assessed using multilevel regression modeling4D-flow MRI, CTPatient-specific 3D geometry (Materialise Mimics)CFD simulations for WSS mapping (EnSight; ANSYS, Inc)Explanted aortic tissue – regional mechanical tensile and peel testingComputational histologyElevated values of WSS were predictive of: reduced wall thickness and dissection energy function (longitudinal).High WSS values also predicted greater ultimate tensile strength.Elevated WSS predicted a reduction in elastin levels and lower abundance of vascular smooth muscle cells.WSS was found to have no effect on collagen abundance or circumferential mechanical properties.4D-flow MRI, Four-dimensional flow-sensitive magnetic resonance imaging; WSS, wall shear stress; OSI, oscillatory shear index; ATAA, ascending thoracic aortic aneurysm; ECG, electrocardiography; CT, computed tomography; BAV, bicuspid aortic valve; FSI, fluid–structure interaction; CFD, computational fluid dynamics; G&R, growth and remodeling. Open table in a new tab 4D-flow MRI, Four-dimensional flow-sensitive magnetic resonance imaging; WSS, wall shear stress; OSI, oscillatory shear index; ATAA, ascending thoracic aortic aneurysm; ECG, electrocardiography; CT, computed tomography; BAV, bicuspid aortic valve; FSI, fluid–structure interaction; CFD, computational fluid dynamics; G&R, growth and remodeling. The distinct differences in blood flow between TAA and normal aortas provide strong evidence for the hypothesis of flow-mediated degeneration. Recent studies, albeit low-powered, have colocalized areas of flow-derived high WSS with focal areas of medial degeneration by analyzing patient-matched explanted aortas.18Bollache E. Guzzardi D.G. Sattari S. Olsen K.E. Di Martino E.S. Malaisrie S.C. et al.Aortic valve-mediated wall shear stress is heterogeneous and predicts regional aortic elastic fiber thinning in bicuspid aortic valve-associated aortopathy.J Thorac Cardiovasc Surg. 2018; 156: 2112-2120Abstract Full Text Full Text PDF PubMed Scopus (94) Google Scholar Further work has linked high WSS with aortic wall thinning, increased wall stiffness, and reduced cellularity4Salmasi M.Y. Pirola S. Sasidharan S. Fisichella S.M. Redaelli A. Jarral O.A. et al.High wall shear stress can predict wall degradation in ascending aortic aneurysms: an integrated biomechanics study.Front Bioeng Biotechnol. 2021; 9: 750656Crossref PubMed Scopus (22) Google Scholar (Figure 2). These associations may reflect the biologic dysfunction driven by WSS-mediated mechanotransduction and regionally variable biomechanical stresses, implying a potential compound effect13Barker A.J. Markl M. Fedak P.W.M. Assessing wall stresses in bicuspid aortic valve-associated aortopathy: forecasting the perfect storm?.J Thorac Cardiovasc Surg. 2018; 156: 471-472Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar: abnormal WSS over time may accelerate wall degeneration, which may be linked with regions of high wall stress (resulting in rupture or dissection).13Barker A.J. Markl M. Fedak P.W.M. Assessing wall stresses in bicuspid aortic valve-associated aortopathy: forecasting the perfect storm?.J Thorac Cardiovasc Surg. 2018; 156: 471-472Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar Despite these theories, as it stands, research groups are yet to establish a causal relationship between abnormal WSS exposure over time and aortic medial degeneration or aortic dissection. An ideal study to establish this would assess baseline WSS in ascending aortas of varying sizes and monitor the outcome (aortic growth, acute events) over the course of time. The ideal clinical decision tool in thoracic aortic disease will contain a diagnostic calculator of aortic wall tension and stress distribution, which will be vital for risk-prediction. In this light, the coupling of fluid motion and solid/wall motion, termed fluid–structure interaction (FSI), is recently emerging as a powerful method in modeling aortic mechanics (in contrast, CFD methods alone often model the aortic wall as a rigid structure, possibly leading to an overestimation of WSS values). The applications of FSI to the cardiovascular system have stemmed from FSI models of engineered systems, such as the behavior of offshore platforms with the ocean, response of tall buildings to wind, stability and response of aircraft wings, and the interaction of dams with reservoirs.22Dowell E.H. Hall K.C. Modeling of fluid-structure interaction.Annu Rev Fluid Mech. 2001; 33: 445-490Crossref Scopus (537) Google Scholar In cardiovascular physiology, the use of FSI was also developed as early as 1972 assessing flow patterns around the aortic valve,23Peskin C.S. Flow patterns around heart valves: a numerical method.J Comput Phys. 1972; 10: 252-271Crossref Scopus (2140) Google Scholar although its application in clinical care never took off. The computational estimation of stresses in the solid domain (aortic wall) usually involves finite element discretization, referred to as “finite element analysis” (FEA). Incorporating the effect of fluid on the solid domain and vice versa, can be assessed in combination and is termed 2-way-FSI. This is more computationally demanding and requires numerous assumptions to be satisfied.24Benra F.K. Dohmen H.J. Pei J. Schuster S. Wan B. A comparison of one-way and two-way coupling methods for numerical analysis of fluid–structure interactions.J Appl Math. 2011; 2011Crossref Scopus (178) Google Scholar The first full FSI simulation of blood flow in the thoracic aorta was performed by Gao and colleagues25Gao F. Watanabe M. Matsuzawa T. Stress analysis in a layered aortic arch model under pulsatile blood flow.Biomed Eng Online. 2006; 5: 25Crossref PubMed Scopus (106) Google Scholar on an idealized 3-layer structure. This work has since been superseded by efforts that have incorporated patient-specific geometries and boundary conditions (Table 1). Most recently, Trabelsi and colleagues16Trabelsi O. Davis F.M. Rodriguez-Matas J.F. Duprey A. Avril S. Patient specific stress and rupture analysis of ascending thoracic aneurysms.J Biomech. 2015; 48: 1836-1843Crossref PubMed Scopus (53) Google Scholar conducted FEA in 5 patients with ascending TAA, demonstrating that peak wall stress was located on the inner curvature of the aneurysm. These results were also confirmed by the same group in 2019,20Condemi F. Campisi S. Viallon M. Croisille P. Avril S. Relationship between ascending thoracic aortic aneurysms hemodynamics and biomechanical properties.IEEE Trans Biomed Eng. 2020; 67: 949-956Crossref PubMed Scopus (18) Google Scholar who applied a fully coupled FSI model using patient-specific geometries and boundary conditions derived from 4D-flow MRI datasets acquired on a single patient. FSI simulations have also been able to model the effect of increased peripheral resistance and increased TAA stiffness on maximal principal stress and the risk of rupture (leading to a dramatic 10-fold increase in stress to 702 kPa).19Campobasso R. Condemi F. Viallon M. Croisille P. Campisi S. Avril S. Evaluation of peak wall stress in an ascending thoracic aortic aneurysm using FSI Simulations: effects of aortic stiffness and peripheral resistance.Cardiovasc Eng Technol. 2018; 9: 707-722Crossref PubMed Scopus (51) Google Scholar Unfortunately, many FEA and FSI studies have to rely on nonpatient-specific arterial wall material properties (including uniform wall thickness).17Trabelsi O. Gutierrez M. Farzaneh S. Duprey A. Avril S. A non-invasive methodology for ATAA rupture risk estimation.J Biomech. 2018; 66: 119-126Crossref PubMed Scopus (15) Google Scholar,20Condemi F. Campisi S. Viallon M. Croisille P. Avril S. Relationship between ascending thoracic aortic aneurysms hemodynamics and biomechanical properties.IEEE Trans Biomed Eng. 2020; 67: 949-956Crossref PubMed Scopus (18) Google Scholar,26Cutugno S. Agnese V. Gentile G. Raffa G.M. Wisneski A.D. Guccione J.M. et al.Patient-specific analysis of ascending thoracic aortic aneurysm with the living heart human model.Bioengineering. 2021; 8: 175Crossref PubMed Scopus (7) Google Scholar As more in vitro studies characterizing ascending aortic wall morphology are published, the incorporation of varying material properties throughout the aortic circumference (including varying thickness) can simulate aortic wall behavior more accurately, including anisotropic deformation patterns and the pathogenesis of aortic dissection. As the research interest grows in this field, the use of FSI will be needed to quantify relationships between WSS variation and the aortic mechanobiological response.21Jamaleddin Mousavi S. Jayendiran R. Farzaneh S. Campisi S. Viallon M. Croisille P. et al.Coupling hemodynamics with mechanobiology in patient-specific computational models of ascending thoracic aortic aneurysms.Comput Methods Programs Biomed. 2021; 205: 106107Crossref PubMed Scopus (17) Google Scholar This includes the effect of flow patterns on aortic remodeling over time (thickness, dilatation, tortuosity) as well as the short-term risk on acute wall failure (ie, dissection/rupture). The spirit of studying aortic biomechanics is to shift the paradigm of clinicians toward the patient-specific underlying mechanisms of disease, moving beyond a surrogate measure of aortic diameter. However, modeling the interaction between blood flow and the arterial wall represents one of the major challenges in the field of patient-specific computational modeling. We are far from determining patient-specific relationships between flow and aortic wall mechanics at present, and further still from developing the numerical algorithms and available software for point of access use in the clinic. Most FSI studies remain very underpowered due to the extreme difficulty in assigning patient-specific material properties of the aortic wall, including variations in wall thickness, tensile properties, central aortic pressure, and precise geometry. Furthermore, increased costs of running such complex simulations limit their availability for the health care setting (FSI models are far more computationally intensive than CFD models). Streamlining computational methods, likely through the use of artificial intelligence, will reduce costs and improve the accuracy of WSS estimation, thus helping to increase the size of research cohorts over the next decade. Ultimately, the future will entail artificial intelligence–based learning systems that synthesize knowledge gained from simulation-based approaches to train algorithms that predict patient-specific risk more accurately.27Liang L. Liu M. Martin C. Elefteriades J.A. Sun W. A machine learning approach to investigate the relationship between shape features and numerically predicted risk of ascending aortic aneurysm.Biomech Model Mechanobiol. 2017; 16: 1519-1533Crossref PubMed Scopus (98) Google Scholar On its own, FEA is not well-suited to simulate failure events. Models that go beyond stress simulation to predict actual material failure, more intricately than conventional FEA, need to be developed in future research. The extended finite element method (ie, xFEM) has been recently shown to simulate aortic rupture and dissection28Wang L. Hill N.A. Roper S.M. Luo X. Modelling peeling- and pressure-driven propagation of arterial dissection.J Eng Math. 2018; 109: 227-238Crossref PubMed Scopus (23) Google Scholar,29Brunet J. Pierrat B. Badel P. A parametric study on factors influencing the onset and propagation of aortic dissection using the extended finite element method.IEEE Trans Biomed Eng. 2021; 68: 2918-2929Crossref PubMed Scopus (7) Google Scholar; however, its use is challenging, particularly in FSI settings. Patient-specific cohort studies that examine the macro- and microstructural effect of flow on the aortic wall (using multimodal approaches) are also important. This will include the downstream response from the endothelium at the cellular level, as well as the adaptation of the aortic wall in terms of continuum mechanics. Such work is only achievable through further interdisciplinary collaboration among clinicians, engineers, and clinical scientists. Without a crucially improved understanding of the effect of aortic flow on wall properties, prognostication for thoracic aortic disease is likely to remain a challenge for the 21st century. The aortic clinic requires more accurate estimates of TAA progression, expansion, and potential dissection. Given the challenges in characterizing TAA pathology (low prevalence, ineffective screening, no measurable symptomatology in chronic state), “going with the flow” toward biomarkers that are congruent with aortic function are crucial for advancing the diagnostic process.