Arteriosclerosis, Thrombosis, and Vascular Biology CLINICAL AND POPULATION STUDIES Low and Oscillatory Wall Shear Stress Is Not Related to Aortic Dilation in Patients With Bicuspid Aortic Valve A Time-Resolved 3-Dimensional Phase-Contrast Magnetic Resonance Imaging Study Lydia Dux-Santoy,* Andrea Guala,* Julio Sotelo, Sergio Uribe, Gisela Teixidó-Turà, Aroa Ruiz-Muñoz, Daniel E. Hurtado, Filipa Valente, Laura Galian-Gay, Laura Gutiérrez, Teresa González-Alujas, Kevin M. Johnson, Oliver Wieben, Ignacio Ferreira-Gonzalez, Arturo Evangelista, José F. Rodríguez-Palomares OBJECTIVE: To assess the relationship between regional wall shear stress (WSS) and oscillatory shear index (OSI) and aortic dilation in patients with bicuspid aortic valve (BAV). APPROACH AND RESULTS: Forty-six consecutive patients with BAV (63% with right-left-coronary-cusp fusion, aortic diameter ≤ 45 mm and no severe valvular disease) and 44 healthy volunteers were studied by time-resolved 3-dimensional phase- contrast magnetic resonance imaging. WSS and OSI were quantified at different levels of the ascending aorta and the aortic arch, and regional WSS and OSI maps were obtained. Seventy percent of BAV had ascending aorta dilation. Compared with healthy volunteers, patients with BAV had increased WSS and decreased OSI in most of the ascending aorta and the aortic Downloaded from http://ahajournals.org by on March 17, 2022 arch. In both BAV and healthy volunteers, regions of high WSS matched regions of low OSI and vice versa. No regions of both low WSS and high OSI were identified in BAV compared with healthy volunteers. Patients with BAV with dilated compared with nondilated aorta presented low and oscillatory WSS in the aortic arch, but not in the ascending aorta where dilation is more prevalent. Furthermore, no regions of concomitant low WSS and high OSI were identified when BAV were compared according to leaflet fusion pattern, despite the well-known differences in regional dilation prevalence. CONCLUSIONS: Regions with low WSS and high OSI do not match those with the highest prevalence of dilation in patients with BAV, thus providing no evidence to support the low and oscillatory shear stress theory in the pathogenesis of proximal aorta dilation in the presence of BAV. VISUAL OVERVIEW: An online visual overview is available for this article. Key Words: aorta ◼ bicuspid aortic valve ◼ magnetic resonance imaging ◼ mechanical stress ◼ regional blood flow ◼ thoracic aortic aneurysm ◼ vascular remodeling P atients with a bicuspid aortic valve (BAV) frequently genetically triggered alterations of the aortic wall and present aortic dilation (60%–84% of cases),1–3 altered hemodynamic stressors have been suggested which is associated with life-threatening complica- to contribute to aortic dilation in BAV.4,5 The recent intro- tions, such as aortic dissection or rupture. Both intrinsic duction of time-resolved 3-dimensional phase-contrast Correspondence to Lydia Dux-Santoy, PhD, Department of Cardiology, Hospital Universitari Vall d´Hebron, Paseo Vall d’Hebron 119-129, 08035, Barcelona, Spain, Email lydia.dux@gmail.com; or Andrea Guala, PhD, Department of Cardiology, Hospital Universitari Vall d´Hebron, Paseo Vall d’Hebron 119-129, 08035, Barcelona, Spain, Email andrea.guala@yahoo.com *L. Dux-Santoy and A. Guala contributed equally to this work and are joint first authors. The online-only Data Supplement is available with this article at https://www.ahajournals.org/doi/suppl/10.1161/ATVBAHA.119.313636. For Sources of Funding and Disclosures, see page e19. © 2019 The Authors. Arteriosclerosis, Thrombosis, and Vascular Biology is published on behalf of the American Heart Association, Inc., by Wolters Kluwer Health, Inc. This is an open access article under the terms of the Creative Commons Attribution Non-Commercial License, which permits use, distribution, and reproduction in any medium, provided that the original work is properly cited and is not used for commercial purposes. Arterioscler Thromb Vasc Biol is available at www.ahajournals.org/journal/atvb e10   January 2020 Arterioscler Thromb Vasc Biol. 2020;40:e10–e20. DOI: 10.1161/ATVBAHA.119.313636 Dux-Santoy et al Oscillating Flow and Bicuspid Aortopathy CLINICAL AND POPULATION Nonstandard Abbreviations and Acronyms Highlights STUDIES - AL 4D flow MRI time-resolved 3-dimensional phase- • This study investigates whether regions of concomi- contrast magnetic resonance imaging tant low wall shear stress and high oscillatory shear AAo ascending aorta index are related to aortic dilation in patients with BAV bicuspid aortic valve bicuspid aortic valve. • Patients with bicuspid aortic valve compared with HV healthy volunteers healthy volunteers did not present regions of low OSI oscillatory shear index and oscillatory shear stress, regardless of the pres- RL-BAV BAV with fusion of right and left ence of dilation. coronary cusps • Concomitant low wall shear stress and high oscillatory RN-BAV BAV with fusion of right coronary shear index did not differentiate patients with bicus- and noncoronary cusps pid aortic valve either by the presence or absence WSS wall shear stress of ascending aorta dilation or by the different cusp- fusion phenotypes despite the well-known differ- ences in regional dilation prevalence between groups. magnetic resonance imaging (4D flow MRI) has per- • These results do not provide evidence to support a mitted substantial advances in the analysis of complex potential role for low and oscillatory shear stress in aortic flow characteristics, particularly in patients with the development of aortic dilation in patients with bicuspid aortic valve. BAV. Although no conclusive data on the role of genet- ics are available,6 several studies have shown that flow alterations may play a role in the development of aortic Study Population aneurysms with different behavior depending on the BAV Patients with BAV with fusion of right and left coronary cusps fusion phenotype.7–9 or right and noncoronary cusps, AAo diameters ≤45 mm and Wall shear stress (WSS) is the most commonly studied no severe valvular disease (aortic regurgitation ≤III, maximum flow parameter associated with aortic dilation; however, its aortic valve velocity <3 m/s) by echocardiography were con- specific role in vascular dilation and aneurysm formation secutively and prospectively recruited between June 2014 and remains unclear. In patients with BAV, a high WSS with an December 2015. Inclusion criteria were age >18 years, no asymmetrical distribution has been associated with degra- hypertension, no connective tissue disorders, no aortic coarcta- Downloaded from http://ahajournals.org by on March 17, 2022 dation of the elastic fibers of the aortic wall.9 From the spa- tion or other congenital heart diseases, and no contraindica- tial point of view, both maximum axial and circumferential tion for cardiac magnetic resonance. Forty-four age-matched healthy volunteers (HV) with a tricuspid aortic valve were stud- components of WSS have been associated with aortopathy ied. The study was approved by the local ethics committee, and in patients with BAV.7,10–12 However, directional variations informed consent was obtained from all participants. in WSS throughout the cardiac cycle, as described by the oscillatory shear index (OSI), have also been related to vas- cular aneurysm formation. Some studies have shown that Cardiac Magnetic Resonance Protocol low WSS and high OSI may activate the atherosclerotic pro- Cardiac magnetic resonance studies were performed on a clini- cess of the wall, leading to dilation of the arterial vessels,13 cal GE 1.5T Signa scanner (General Electric, WI). The protocol included 2-dimensional balanced steady-state free-precession at least in intracranial aneurysm formation.14 This mecha- cine images, which were used to assess BAV fusion pheno- nism has also been suggested to contribute to ascending type and aortic diameters (using the double-oblique cine car- aorta (AAo) dilation in patients with tricuspid aortic valve.15,16 diac magnetic resonance at late diastole) and a time-resolved However, whether this mechanism may contribute to aor- 3-dimensional phase-contrast MRI acquisition (also known tic dilation in patients with BAV remains to be established. as 4D flow MRI) with retrospective ECG gating during free- Indeed, although WSS has been described in the AAo in breathing. Endovenous contrast was not given. several 4D flow MRI studies and related to aortic dilation in For 4D flow MRI, data were acquired using the VIPR (vastly patients with BAV,7,8,10 the role of OSI has been evaluated in undersampled isotropic projection) sequence, a radially under- very few studies with limited populations.17,18 sampled acquisition with 5-point balanced velocity encoding.19 Thus, using 4D flow MRI, this study aimed to analyze The acquisition volume included the entire thoracic aorta. regional abnormalities in aortic WSS and OSI to ascer- Acquisitions were made with an 8-channel cardiac coil using the following parameters: velocity encoding 200 cm/s, field of tain whether low and oscillatory shear stress is locally view 400×400×400 mm, scan matrix 160×160×160 (voxel associated with aortic dilation in these patients. size 2.5×2.5×2.5 mm), flip angle 8º, repetition time 4.2 to 7.2 ms and echo time 2.2 to 3.6 ms. This data set was reconstructed according to the nominal temporal resolution of each patient MATERIALS AND METHODS and was (5×repetition time) 21 to 36 ms. Reconstructions The data that support the findings of this study are available were performed offline with corrections for background phase, from the corresponding author upon reasonable request. eddy currents and trajectory errors.19 Arterioscler Thromb Vasc Biol. 2020;40:e10–e20. DOI: 10.1161/ATVBAHA.119.313636 January 2020   e11 Dux-Santoy et al Oscillating Flow and Bicuspid Aortopathy Aortic Dilation were obtained by averaging over each of the 8 circumferential CLINICAL AND POPULATION To determine the presence of aortic dilation in patients with segments (Figure 1C) and were used to create coregistered BAV, maximum diameters at aortic root and mid AAo were WSS and OSI maps, as previously described.7,12 Maps show- STUDIES - AL adjusted with a logarithm transformation to calculate the Z ing regions of low and oscillatory WSS were obtained as the score accounting for sex, age, and body surface area.20 Aortic overlap between regions of low WSS and regions of high OSI. dilation was considered when at least one Z score was over 2.3 Statistical Analysis WSS and OSI: Computation and Analysis Continuous variables were expressed as mean±SD. The WSS magnitude and OSI maps were computed from 4D Shapiro-Wilk test was used to evaluate normal distribution. flow acquisitions by applying a 3-dimensional finite-element Differences between groups for continuous parameters were method, as previously described.21,22 Briefly, the thoracic assessed by Student t test if they presented a normal distribu- aorta was semiautomatically segmented and used to gener- tion and Mann-Whitney U test otherwise. χ2 test was applied ate a tetrahedral mesh using the iso2mesh Matlab toolbox23 for categorical variables, which were reported as percentages. and Computational Geometry Algorithms Library (http:// A 2-tailed P<0.05 was considered statistically significant. www.cgal.org). Using cubic interpolation, velocity values SPSS 19.0 software version (IBM SPSS Statistics, IL) was were transferred from the 4D flow data to each node of the used for the analysis. generated mesh. This flow field was later analyzed in terms of local, spatial, and temporal derivatives to compute WSS magnitude and OSI. RESULTS Three anatomic reference points (sinotubular junc- tion, brachiocephalic trunk, and left subclavian artery) were Demographics located in the AAo and the aortic arch using 4D flow-derived, Forty-six patients with BAV (63% right and left coronary phase-contrast enhanced magnetic resonance angiogra- cusp phenotype, 70% with AAo dilation) and 44 HV com- phy (Figure 1A). The aortic centerline was automatically computed for each segmented aorta using The Vascular pleted the study protocol. HV and BAV were matched in Modeling Toolkit24 (VMTK, Orobix, Bergamo, Italy; www.vmtk. terms of age, sex, body surface area, and blood pres- org) and used to calculate 10 double-oblique analysis planes sure (Table). As expected, BAV had higher diameter (orthogonal to the centerline) equally distributed in 2 analysis and Z score at both the aortic root and AAo. Compared regions: 6 in the AAo, between the sinotubular junction and with HV, nondilated BAV presented statistically signifi- the origin of the brachiocephalic trunk, and 4 in the aortic cant differences only in aortic root and AAo diameters Downloaded from http://ahajournals.org by on March 17, 2022 arch between the brachiocephalic trunk and the left subcla- (P=0.039 and P=0.002, respectively) and AAo Z score vian artery (Figure 1B). Each region was further divided into (P<0.0001). BAV with fusion of right and left coronary 8 equal segments along the aortic circumference: outer, left- cusps (RL-BAV) and BAV with fusion of right coronary outer, left, left-inner, inner, right-inner (RI), right, right-outer and noncoronary cusps (RN-BAV) patients had similar (RO; Figure 1C). Peak-systolic WSS and OSI values of the contour lumen demographic and clinical conditions, but RN-BAV pre- were interpolated using a cubic spline approximation to 80 sented higher systolic blood pressure (P=0.034). No dif- points uniformly distributed over each contour. Contour- ferences other than aortic diameters and Z scores were averaged WSS and OSI were calculated, respectively, as the obtained when comparing patients with nondilated and mean values of WSS and OSI in each contour. Regional values dilated BAV. Figure 1. Graphical representation of the analysis regions. A, Acquired volume containing the segmentation of the aorta and the 3 anatomic reference points: sinotubular junction (blue), the brachiocephalic trunk bifurcation (red), and the left subclavian artery (green). B, Using the anatomic references, 6 double-oblique planes (planes 1–6) were distributed in the ascending aorta and 4 (planes 7–10) in the aortic arch. C, The aorta was circumferentially divided into 8 segments: outer (O), left-outer (LO), left (L), left-inner (LI), inner (I), right-inner (RI), right (R), right-outer (RO). e12   January 2020 Arterioscler Thromb Vasc Biol. 2020;40:e10–e20. DOI: 10.1161/ATVBAHA.119.313636 Dux-Santoy et al Oscillating Flow and Bicuspid Aortopathy Table. Demographics and Aortic Dimensions in HV and Patients With BAV CLINICAL AND POPULATION BAV Groups STUDIES - AL BAV Phenotype BAV Dilation HV BAV P Value RL-BAV RN-BAV P Value NON-DIL DIL P Value N 44 46 29 17 14 32 Age, y 43±13 45±14 0.404 45±12 46±19 0.795 45±16 46±14 0.865 Men, % 57 59 0.857 62 53 0.544 79 50.0 0.070 BSA, m2 1.8±0.2 1.8±0.2 0.368 1.8±0.2 1.8±0.1 0.678 1.8±0.2 1.8±0.2 0.400 IAo, % … 0.192 0.103 None … 30.4 27.6 35.3 28.6 31.3 Mild … 50.0 62.1 35.3 42.8 40.6 Moderate … 19.6 10.3 29.4 28.6 28.1 EAo, % … 0.052 0.332 None … 93.5 96.6 88.2 100.0 90.6 Mild … 6.5 3.4 11.3 0.0 9.4 SBP, mm Hg 130±19 136±19 0.113 132±17 144±20 0.034 134±17 137±19 0.555 DBP, mm Hg 73±11 76±9 0.125 75±7 79±10 0.105 77±7 76±9 0.861 Diameter SoV, mm 30.5±3.8 35.3±5.1 <0.001 35.4±5.9 35.1±3.4 0.851 33.0±3.6 36.3±5.3 0.040 Diameter AAo, mm 28.2±3.8 38.4±7.2 <0.001 38.0±7.5 39.0±6.7 0.646 32.3±4.1 41.1±6.6 <0.001 Z score SoV −0.3±1.1 1.21±1.4 <0.001 1.2±1.3 1.2±1.6 0.906 0.3±0.8 1.6±1.4 0.002 Z score AAo −0.2±0.9 2.7±1.7 <0.001 2.6±1.7 2.9±1.6 0.525 1.0±1.1 3.5±1.3 <0.001 Patients with BAV are compared with HV and according to their fusion phenotype and dilation. Results are presented as mean±SD. P value for each comparison are included. AAo indicates ascending aorta; BAV, bicuspid aortic valve; BSA, body surface area; DBP, diastolic blood pressure; DIL, dilated; EAo, aortic stenosis; HV, healthy volunteers; IAo, aortic insufficiency; RL-BAV, BAV with fusion of right and left coronary cusps; RN-BAV, BAV with fusion of right coronary and noncoronary cusps; SBP, systolic blood pressure; and SoV, sinus of Valsalva. Downloaded from http://ahajournals.org by on March 17, 2022 Contour-Averaged WSS and OSI statistically significant differences were obtained when patients with BAV were compared according to the pres- Contour-averaged WSS magnitude (at peak systole) and ence of aortic dilation (Figure 3B). Nonetheless, both OSI were compared between BAV and HV (Figure 2) WSS and OSI tended to be lower in the AAo of patients and in patients with BAV according to the cusp-fusion with BAV with a dilated aorta, whereas lower WSS but and the presence of dilation (Figure 3). higher OSI was present in the aortic arch. Differences Between Patients With BAV and HV Results of the comparison between BAV and HV, con- sidering the presence of aortic dilation in BAV, are sum- Regional WSS and OSI marized in Figure 2. Figure 4 shows regional WSS and OSI maps in HV and Patients with BAV compared with HV presented patients with BAV. Moreover, WSS and OSI maps in the significantly higher contour-averaged WSS magnitude subgroups of patients with BAV with and without aor- (Figure 2A, left) but lower OSI (Figure 2A, right) along tic dilation as well as in patients with BAV according to the whole AAo and the aortic arch. Similar results were the leaflet fusion pattern are presented. Corresponding obtained when HV with patients with nondilated (Fig- maps of regional statistical differences (P<0.05) in WSS ure 2B) and dilated BAV were compared (Figure 2C). and OSI in the comparison between BAV and HV and In patients with dilated BAV, however, WSS was slightly when patients with BAV were compared according to decreased in the proximal aortic arch and did not differ fused cusps and dilation are shown in Figure 5 and Fig- from HV in this region (Figure 2C, left). ure 6, respectively. For all comparisons, the presence of regions with concomitant low WSS and high OSI is also Differences in Patients With BAV According to reported in the right panels of Figure 5 and Figure 6. Valvular Phenotype and Dilation When patients with BAV were compared according to Differences Between Patients With BAV and HV the fused cusps, patients with RL-BAV presented higher In patients with BAV, regional WSS magnitude showed an WSS than RN-BAV in the proximal AAo (Figure 3A, asymmetrical distribution, with maximum WSS extending left), with no differences in OSI in this region (Figure 3A, from the RI wall in the proximal AAo to the RO wall in right). Conversely, higher OSI, but similar WSS, was pres- the distal AAo (Figure 4B, left). This asymmetry was also ent in the distal AAo and the proximal aortic arch. No present in the aortic arch of BAV but was notably lower Arterioscler Thromb Vasc Biol. 2020;40:e10–e20. DOI: 10.1161/ATVBAHA.119.313636 January 2020   e13 Dux-Santoy et al Oscillating Flow and Bicuspid Aortopathy CLINICAL AND POPULATION STUDIES - AL Downloaded from http://ahajournals.org by on March 17, 2022 Figure 2. Contour-averaged wall shear stress (WSS) and oscillatory shear index (OSI) comparing patients with bicuspid aortic valve (BAV) and healthy volunteers (HV). Comparison is performed in the ascending aorta (planes 1-6) and the aortic arch (planes 7-10) of (A) patients with BAV vs HV; (B) nondilated BAV vs HV; (C) dilated BAV vs HV. DIL indicates dilated ascending aorta. *Statistically significant difference (P<0.05). owing to the reduced WSS in this region. By contrast, HV Differences in Patients With BAV According to presented homogenously distributed low WSS in the AAo Valvular Phenotype and Dilation and the aortic arch (Figure 4A, left). Regional OSI maps Regional WSS and OSI were also analyzed in patients showed a relative uniform distribution of OSI in HV, which with BAV according to the leaflet fusion pattern (Fig- was slightly higher in the left-inner wall along the AAo ure 4E, Figure 4F, and Figure 6B) and the presence or and the aortic arch (Figure 4A, right). Conversely, BAV absence of aortic dilation (Figure 4C, Figure 4D, and showed lower OSI compared with HV, with maximum val- Figure 6A). ues extending from the RI mid AAo to RO wall of the Regional maps presented differences in the exten- aortic arch (Figure 4B, right). Interestingly, regions with sion of high WSS in the different BAV phenotypes, with lower WSS values matched those with higher OSI and a maximum WSS extending from the RI wall of the proxi- vice versa both in HV (Figure 4A) and BAV (Figure 4B). mal AAo to the RO wall of the mid AAo in RL-BAV (Fig- Differences in WSS and OSI between HV and BAV ure 4E) and from the right wall of the mid AAo to the RO emerged in the regional statistical significance maps, wall in the distal AAo in RN-BAV (Figure 4F). OSI was where patients with BAV presented higher WSS and locally increased only in the distal aortic arch in patients lower OSI in most of the AAo and the aortic arch (Fig- with RN-BAV. Conversely, in patients with RL-BAV, OSI ure 5A). Thus, regions of concomitant high WSS and low was high along the RI wall of the AAo and the RI to RO OSI were obtained in BAV compared with HV (Figure 5A, wall of the aortic arch (Figure 4F and Figure 4E, respec- right). These results were confirmed when patients with tively). Despite these differences, statistical significance nondilated (Figure 4C and Figure 5B) and dilated BAV was reached in some regions (Figure 6B, left), but no were compared (Figure 4D and Figure 5C) with HV. regions with concomitant low WSS and high OSI were In summary, no regions of concomitant low WSS and obtained (Figure 6B, right). high OSI were present in BAV compared with HV, regard- Regions of concomitant low WSS and high OSI were less of the presence of dilation. only obtained in patients with dilated BAV compared with e14   January 2020 Arterioscler Thromb Vasc Biol. 2020;40:e10–e20. DOI: 10.1161/ATVBAHA.119.313636 Dux-Santoy et al Oscillating Flow and Bicuspid Aortopathy CLINICAL AND POPULATION STUDIES - AL Figure 3. Contour-averaged wall shear stress (WSS) and oscillatory shear index (OSI) in patients with bicuspid aortic valve (BAV) according to their fused cusps and aortic dilation. Comparison is performed in the ascending aorta (planes 1-6) and the aortic arch (planes 7-10) of (A) RL-BAV vs RN-BAV patients and (B) BAV patients with nondilated vs dilated aorta. DIL indicates dilated ascending aorta; RL-BAV, BAV with fusion of right and left coronary cusps; RN-BAV, BAV with fusion of right coronary and noncoronary cusps. *Statistically significant difference (P<0.05). nondilated and located in a limited area in the RO wall of process13,27,28 in the cerebrovascular circulation14,29 and the aortic arch (Figure 6A, right). in the AAo of patients with tricuspid aortic valve.16 Thus, Downloaded from http://ahajournals.org by on March 17, 2022 it is of great interest to assess whether low and oscilla- tory WSS may also contribute to aortic dilation in BAV. DISCUSSION This study evaluates whether this mechanism is related The main findings of our study were (1) patients with to aortic dilation in patients with BAV. BAV did not present regions of concomitant low WSS and high OSI compared with HV, regardless of the WSS and OSI in Patients With BAV Compared presence of dilation; (2) concomitant low WSS and high OSI in the AAo did not differentiate dilated from With HV nondilated patients with BAV or patients with RL-BAV In line with previous studies,8,30 regional WSS magni- from patients with RN-BAV despite the well-known dif- tude was higher in the RI to RO AAo of patients with ference in regional dilation prevalence between these BAV compared with HV, regardless of aortic dilation groups; and (3) regions of high OSI matched regions (Figure 5). In HV, the highest OSI was found along the of low WSS magnitude and vice versa. Taken together, left-inner region of AAo and aortic arch, thus matching these results do not provide evidence to support a with the region where flow tends to separate from the potential role for low and oscillatory WSS in the devel- aortic wall.31 By contrast, in patients with BAV, OSI was opment of BAV aortopathy. higher in the RI to RO wall of the distal AAo and aortic Aortic flow alterations are one of the most important arch (Figure 4B), a result that may be explained by the factors related to aneurysm formation in patients with asymmetrical velocity profile characteristic of patients BAV.7,10,12,25,26 Building from previous studies relating BAV with BAV. No evidence to sustain the low and oscillatory and cusp-fusion phenotypes to regional dilation, this study WSS theory was found in this study. Indeed, no regions assessed whether low and oscillatory WSS is related to with concomitant low WSS and high OSI were observed aortic dilation in BAV. Previous studies related aortic dila- in patients with BAV compared with HV, while the oppo- tion in BAV to high asymmetrical shear stresses result- site was true (Figure 5). These results may be partially ing from flow eccentricity7,8 and abnormally increased explained by the strong, inverse relationship between rotational flow.7,12 However, other studies showed low (contour-averaged and regional) WSS and OSI (Fig- and oscillatory WSS to be related to vascular dilation and ure 4), as previously observed in a small study17 and aneurysm formation through an atherosclerosis-driven confirmed here. Arterioscler Thromb Vasc Biol. 2020;40:e10–e20. DOI: 10.1161/ATVBAHA.119.313636 January 2020   e15 Dux-Santoy et al Oscillating Flow and Bicuspid Aortopathy CLINICAL AND POPULATION STUDIES - AL Downloaded from http://ahajournals.org by on March 17, 2022 Figure 4. Regional wall shear stress (WSS) and oscillatory shear index (OSI) maps in the ascending aorta and the aortic arch of healthy volunteers (HV) and patients with bicuspid aortic valve (BAV). HV (A); patients with BAV (B); BAV with nondilated aorta (C); BAV with dilated aorta (D); BAV with fusion of right and left coronary cusps (­R L- BAV; E); BAV with fusion of right coronary and noncoronary cusps (RN-BAV; F). I indicates inner; L, left; O, outer; and R, right. e16   January 2020 Arterioscler Thromb Vasc Biol. 2020;40:e10–e20. DOI: 10.1161/ATVBAHA.119.313636 Dux-Santoy et al Oscillating Flow and Bicuspid Aortopathy CLINICAL AND POPULATION STUDIES - AL Downloaded from http://ahajournals.org by on March 17, 2022 Figure 6. Regional differences in wall shear stress (WSS) and oscillatory shear index (OSI), and regions with concomitant low WSS and high OSI when comparing patients with bicuspid aortic valve (BAV) according to their fused cusps and aortic dilation. BAV with nondilated vs dilated aorta (A), and BAV with fusion of right and left coronary cusps (RL-BAV; B) vs BAV with fusion of right coronary and noncoronary cusps (RN-BAV). AAo indicates ascending aorta; I, inner; L, left; O, outer; and R, right. Figure 5. Regional differences in wall shear stress (WSS) and thus present in the RO aortic arch of dilated versus non- oscillatory shear index (OSI), and regions with concomitant dilated BAV. Bürk et al16 compared patients with tricuspid low WSS and high OSI when comparing healthy volunteers (HV) and patients with bicuspid aortic valve (BAV). aortic valve according to AAo dilation and obtained simi- BAV vs HV (A); BAV with nondilated aorta vs HV (B); and BAV with lar results, thus suggesting a potential role of low WSS dilated aorta vs HV (C). The legend is common for all panels. AAo and high OSI in aortic dilation. Because increased OSI in indicates ascending aorta; I, inner; L, left; O, outer; and R, right. the aortic arch is observed in both patients with bicuspid and tricuspid aortic valve only when dilation is present, WSS and OSI in Patients With BAV According to this characteristic may well be a consequence of dilation. Dilation Of note, our data do not concur with a small study in 5 patients with BAV, which reported increased OSI in non- Consecutive patients with BAV presented higher OSI dilated BAV.17 In this context, it is important to underline in the RI to RO wall of the distal AAo compared with that flow in the aortic arch is impacted by aortic curva- HV. This extends to the very proximal aortic arch in non- ture and the presence of supra-aortic trunks.28,32 These dilated BAV (Figure 5B, left) and all along the arch in characteristics may promote local recirculation zones,33,34 dilated BAV (Figure 5C, left and Figure 6A, left). A small thus creating regions of low and oscillatory WSS12,32 region with both low WSS magnitude and high OSI was that may contribute to aortic arch aneurysm formation.28 Arterioscler Thromb Vasc Biol. 2020;40:e10–e20. DOI: 10.1161/ATVBAHA.119.313636 January 2020   e17 Dux-Santoy et al Oscillating Flow and Bicuspid Aortopathy However, considering our results and previous investiga- quality. Thus, this study cannot rule out a possible role of CLINICAL AND POPULATION tions, we conclude that there is not sufficient evidence OSI in patients with BAV with severe valvular disease, to support the contribution of this mechanism to aortic larger aortic diameters, or infrequent aortic root dilation. STUDIES - AL dilation in patients with BAV. The number of patients with BAV (46) and HV (44) included in the study is limited, and particularly affects comparisons among BAV subgroups (dilation and val- WSS and OSI in Patients With BAV According to vular phenotype). However, the reliability of the quanti- Cusp-Fusion Phenotype tative description of flow characteristics from 4D flow Maximum regional WSS was obtained in the RI to RO MRI shown in previous studies based on similar or even wall of the proximal-mid AAo in RL-BAV and in the RO smaller groups.8,10,25 should guarantee that the study was distal AAo in RN-BAV (Figure 4E and Figure 4F, respec- powered to detect eventual hemodynamic alterations, as tively), matching the outflow jet direction.7 Regions of indeed shown for WSS. relatively high OSI were observed in the mid AAo inner Contour-averaging of shear stress measurements wall of RL-BAV and in the proximal AAo left area of leads to the removal of the spatial distribution. To provide patients with RN-BAV. Of note, these regions showed regional distribution information of shear stress, all com- correspondence with the areas of minimum WSS and parisons in the study have been also proposed through did not match areas with different prevalence of dila- regional WSS and OSI maps. tion between BAV phenotypes (proximal AAo in RL-BAV The aortic wall was described using a common seg- and distal AAo and proximal aortic arch in RN-BAV4,12). mentation for all cardiac phases. Thus, the calculation of Furthermore, no regions of low and oscillatory WSS OSI does not account for the aortic movement along the were identified when BAV phenotypes were compared cardiac cycle. Furthermore, results in the aortic arch may (Figure 6B, right). These results further downgrade the be affected by the heterogeneous location of the supra- hypothesis of an etiological role of low and oscillatory aortic trunks and aortic curvature. However, as all acqui- WSS for aortic dilation in patients with BAV. sitions were made using the same imaging parameters and analyzed with the same methods, the comparison between groups was unbiased. Flow Alterations and Dilation in Patients With BAV Conclusions Downloaded from http://ahajournals.org by on March 17, 2022 Low and oscillatory WSS has been shown to play a role in atherosclerotic aneurysms14 and may contribute No regions of concomitant low WSS magnitude and high to thrombus deposition and abdominal aortic dilation.35 OSI were observed in patients with BAV compared with However, our study suggests that this mechanism does HV. Although distinct BAV phenotypes are related to dif- not play a role in aortic dilation in BAV. BAV aortic dila- ferent regions of predominant dilation, their comparison tion has been previously related to increased flow dis- in our study did not show regions of concomitant low placement,8 increased and asymmetrically distributed WSS magnitude and high OSI. These results provide no WSS,7,9,36 and increased in-plane and through-plane evidence to support the contribution of low and oscilla- rotational flow.7,12 These flow alterations contribute tory shear stress measures to the pathogenesis of aortic to extracellular matrix dysregulation and elastic fiber dilation in patients with BAV. degeneration, thereby promoting aortic remodeling.9 The differences in aortic wall structure at different locations ARTICLE INFORMATION in the aorta and in wall structure, shape, hemodynamic Received August 8, 2019; accepted November 11, 2019. environment, and acting pressure between the aorta and intracranial arteries may well account for a different Affiliations mechanism in aneurysms etiology. From the Department of Cardiology, CIBERCV, Universitat Autònoma de Barcelona, Vall d’Hebron Institut de Recerca (VHIR), Hospital Universitari Vall d´Hebron, Bar- celona, Spain (L.D.-S., A.G., G.T.-.T., A.R.-M., F.V., L.G.-.G., L.G., T.G.-A., A.E., J.F.R.P.); Department of Cardiology, CIBERESP, Universitat Autònoma de Barcelona, Vall Limitations d'Hebron Institut de Recerca (VHIR), Hospital Universitari Vall d´Hebron, Barcelona, The cross-sectional nature of the present study pre- Spain (I.F.-G.); Biomedical Imaging Center (J.S., S.U.), Department of Electrical En- gineering, School of Engineering (J.S.), Department of Radiology, School of Medi- cludes the investigation of eventual causal relationships. cine (S.U.), Institute for Biological and Medical Engineering, Schools of Engineer- Further longitudinal studies are needed to confirm the ing, Medicine, and Biological Sciences (D.E.H.), and Department of Structural and results of this study and investigate causal relationships. Geotechnical Engineering, Schools of Engineering (D.E.H.), Pontificia Universidad Católica de Chile, Santiago; Millennium Nucleus for Cardiovascular Magnetic Reso- Only patients with BAV with nonsevere valvular dys- nance, Santiago, Chile (J.S., S.U., D.E.H.); and Department of Medical Physics (K.M.J., function and aortic diameters ≤45 mm were included O.W.) and Department of Radiology (K.M.J., O.W.), University of Wisconsin-Madison. to limit possible hemodynamic alterations secondary to Acknowledgments aortic dilation or severe valvular disease. Moreover, the We are grateful to Augusto Sao Avilés for statistical analysis and Christine O’Hara aortic root was excluded because of poor segmentation for English revisions. e18   January 2020 Arterioscler Thromb Vasc Biol. 2020;40:e10–e20. DOI: 10.1161/ATVBAHA.119.313636 Dux-Santoy et al Oscillating Flow and Bicuspid Aortopathy Sources of Funding 12. Dux-Santoy L, Guala A, Teixidó-Turà G, Ruiz-Muñoz A, Maldonado G, CLINICAL AND POPULATION This study has been funded by Spanish Ministry of Economy and Competi- Villalva N, Galian L, Valente F, Gutiérrez L, González-Alujas T, et al. Increased tiveness (RTC-2016-5152-1), Instituto de Salud Carlos III (PI14/0106 and rotational flow in the proximal aortic arch is associated with its dilation in bicuspid aortic valve disease. Eur Hear J - Cardiovasc Imaging. 2019;34:1– STUDIES - AL PI17/00381), La Marató de TV3 (project number 20151330) and Centro de Investigación Biomédica en Red - Enfermedades Cardiovasculares (CIBERCV). 11. doi:10.1093/ehjci/jez046 A. Guala has received funding from the European Union Seventh Framework 13. Malek AM, Alper SL, Izumo S. Hemodynamic shear stress and its role Programme FP7/People (267128). J. Sotelo, D.E. Hurtado, and S. Uribe were in atherosclerosis. JAMA. 1999;282:2035–2042. doi: 10.1001/jama. funded by the Millenium Science Initiative of the Ministry of Economy, Develop- 282.21.2035 ment and Tourism, Grant Nucleus for Cardiovascular Magnetic Resonance and 14. Meng H, Tutino VM, Xiang J, Siddiqui A. High WSS or low WSS? Com- projects CONICYT-FONDECYT No. 1181057, Comisión Nacional de Investig- plex interactions of hemodynamics with intracranial aneurysm initiation, ación Científica y Tecnológica - Fondo Nacional de Desarrollo Científico y Tec- growth, and rupture: toward a unifying hypothesis. AJNR Am J Neuroradiol. nológico (CONICYT-FONDECYT) Postdoctorado 2017 No. 3170737. 2014;35:1254–1262. doi: 10.3174/ajnr.A3558 15. Frydrychowicz A, Stalder AF, Russe MF, Bock J, Bauer S, Harloff A, Berger A, Disclosures Langer M, Hennig J, Markl M. Three-dimensional analysis of segmental wall K.M. Johnson has received a research grant from GE Healthcare and from Myo- shear stress in the aorta by flow-sensitive four-dimensional-MRI. J Magn cardial Solutions through The University of Wisconsin-Madison (no personal com- Reson Imaging. 2009;30:77–84. doi: 10.1002/jmri.21790 pensation) and holds a consulting agreement with Vertex Pharmaceuticals. O. 16. Bürk J, Blanke P, Stankovic Z, Barker A, Russe M, Geiger J, Wieben has received a research grant from GE Healthcare through The Univer- Frydrychowicz A, Langer M, Markl M. Evaluation of 3D blood flow pat- sity of Wisconsin-Madison (no personal compensation). The other authors report terns and wall shear stress in the normal and dilated thoracic aorta using no conflicts. flow-sensitive 4D CMR. J Cardiovasc Magn Reson. 2012;14:84. doi: 10.1186/1532-429X-14-84 17. Piatti F, Sturla F, Bissell MM, Pirola S, Lombardi M, Nesteruk I, Della Corte A, Redaelli ACL, Votta E. 4D flow analysis of BAV-related fluid-dynamic altera- REFERENCES tions: evidences of wall shear stress alterations in absence of clinically- 1. Nkomo VT, Enriquez-Sarano M, Ammash NM, Melton LJ 3rd, Bailey KR, relevant aortic anatomical remodeling. Front Physiol. 2017;8:441. doi: Desjardins V, Horn RA, Tajik AJ. Bicuspid aortic valve associated with 10.3389/fphys.2017.00441 aortic dilatation: a community-based study. Arterioscler Thromb Vasc Biol. 18. Barker AJ, Lanning C, Shandas R. Quantification of hemodynamic wall 2003;23:351–356. doi: 10.1161/01.atv.0000055441.28842.0a shear stress in patients with bicuspid aortic valve using phase-contrast MRI. 2. Michelena HI, Khanna AD, Mahoney D, Margaryan E, Topilsky Y, Ann Biomed Eng. 2010;38:788–800. doi: 10.1007/s10439-009-9854-3 Suri RM, Eidem B, Edwards WD, Sundt TM 3rd, Enriquez-Sarano M. Inci- 19. Johnson KM, Markl M. Improved SNR in phase contrast velocimetry with dence of aortic complications in patients with bicuspid aortic valves. JAMA. five-point balanced flow encoding. Magn Reson Med. 2010;63:349–355. 2011;306:1104–1112. doi: 10.1001/jama.2011.1286 doi:10.1002/mrm.22202 3. Evangelista A, Gallego P, Calvo-Iglesias F, Bermejo J, Robledo-Carmona J, 20. Campens L, Demulier L, De Groote K, Vandekerckhove K, De Wolf D, Sánchez V, Saura D, Arnold R, Carro A, Maldonado G, et al. Anatomical and clin- Roman MJ, Devereux RB, De Paepe A, De Backer J. Reference values ical predictors of valve dysfunction and aortic dilation in bicuspid aortic valve for echocardiographic assessment of the diameter of the aortic root and disease. Heart. 2018;104:566–573. doi:10.1136/heartjnl-2017-311560 ascending aorta spanning all age categories. Am J Cardiol. 2015;114:914– 4. Verma S, Siu SC. Aortic dilatation in patients with bicuspid aortic valve. N 920. doi:10.1016/j.amjcard.2014.06.024 21. Sotelo J, Urbina J, Valverde I, Tejos C, Irarrazaval P, Andia ME, Uribe S, Downloaded from http://ahajournals.org by on March 17, 2022 Engl J Med. 2014;370:1920–1929. doi: 10.1056/NEJMra1207059 5. Guala A, Rodriguez-Palomares J, Dux-Santoy L, Teixido-Tura G, Maldonado G, Hurtado DE. 3D quantification of wall shear stress and oscillatory shear Galian L, Huguet M, Valente F, Gutiérrez L, González-Alujas T, et al. Influ- index using a finite-element method in 3D CINE PC-MRI data of the ence of aortic dilation on the regional aortic stiffness of bicuspid aortic valve thoracic aorta. IEEE Trans Med Imaging. 2016;35:1475–1487. doi: assessed by 4-dimensional flow cardiac magnetic resonance: comparison 10.1109/TMI.2016.2517406 with marfan syndrome and degenerative aortic aneurysm. JACC Cardiovasc 22. Sotelo J, Dux-Santoy L, Guala A, Rodríguez-Palomares J, Evangelista A, Imaging. 2019;12:1020–1029. doi: 10.1016/j.jcmg.2018.03.017 Sing-Long C, Urbina J, Mura J, Hurtado DE, Uribe S. 3D axial and circum- 6. Prakash SK, Bossé Y, Muehlschlegel JD, Michelena HI, Limongelli G, ferential wall shear stress from 4D flow MRI data using a finite element Della Corte A, Pluchinotta FR, Russo MG, Evangelista A, Benson DW, et al; method and a laplacian approach. Magn Reson Med. 2018;79:2816–2823. BAVCon Investigators. A roadmap to investigate the genetic basis of bicus- doi: 10.1002/mrm.26927 pid aortic valve and its complications: insights from the international BAVCon 23. Qianqian F, Boas DA. Tetrahedral mesh generation from volumetric binary (Bicuspid Aortic Valve Consortium). J Am Coll Cardiol. 2014;64:832–839. and gray-scale images. Proc IEEE Int Symp Biomed Imag. 2009:1142–1145. doi: 10.1016/j.jacc.2014.04.073 24. Antiga L, Piccinelli M, Botti L, Ene-Iordache B, Remuzzi A, Steinman DA. 7. Rodríguez-Palomares JF, Dux-Santoy L, Guala A, Kale R, Maldonado G, An image-based modeling framework for patient-specific computa- Teixidó-Turà G, Galian L, Huguet M, Valente F, Gutiérrez L, et al. Aortic flow tional hemodynamics. Med Biol Eng Comput. 2008;46:1097–1112. doi: patterns and wall shear stress maps by 4D-flow cardiovascular magnetic res- 10.1007/s11517-008-0420-1 onance in the assessment of aortic dilatation in bicuspid aortic valve disease. 25. Meierhofer C, Schneider EP, Lyko C, Hutter A, Martinoff S, Markl M, Hager A, J Cardiovasc Magn Reson. 2018;20:28. doi: 10.1186/s12968-018-0451-1 Hess J, Stern H, Fratz S. Wall shear stress and flow patterns in the ascend- 8. Mahadevia R, Barker AJ, Schnell S, Entezari P, Kansal P, Fedak PW, ing aorta in patients with bicuspid aortic valves differ significantly from tri- Malaisrie SC, McCarthy P, Collins J, Carr J, et al. Bicuspid aortic cusp fusion cuspid aortic valves: a prospective study. Eur Heart J Cardiovasc Imaging. morphology alters aortic three-dimensional outflow patterns, wall shear 2013;14:797–804. doi: 10.1093/ehjci/jes273 stress, and expression of aortopathy. Circulation. 2014;129:673–682. doi: 26. Guala A, Rodriguez-Palomares J, Galian-Gay L, Teixido-Tura G, Johnson KM, 10.1161/CIRCULATIONAHA.113.003026 Wieben O, Sao Avilés A, Evangelista A. Partial aortic valve leaflet fusion is 9. Guzzardi DG, Barker AJ, van Ooij P, Malaisrie SC, Puthumana JJ, related to deleterious alteration of proximal aorta hemodynamics. Circulation. Belke DD, Mewhort HE, Svystonyuk DA, Kang S, Verma S, et al. Valve- 2019;139:2707–2709. doi: 10.1161/CIRCULATIONAHA.119.039693 related hemodynamics mediate human bicuspid aortopathy: insights from 27. Peiffer V, Sherwin SJ, Weinberg PD. Does low and oscillatory wall shear wall shear stress mapping. J Am Coll Cardiol. 2015;66:892–900. doi: stress correlate spatially with early atherosclerosis? A systematic review. 10.1016/j.jacc.2015.06.1310 Cardiovasc Res. 2013;99:242–250. doi: 10.1093/cvr/cvt044 10. Bissell MM, Hess AT, Biasiolli L, Glaze SJ, Loudon M, Pitcher A, 28. Cecchi E, Giglioli C, Valente S, Lazzeri C, Gensini GF, Abbate R, Mannini L. Davis A, Prendergast B, Markl M, Barker AJ, et al. Aortic dilation in bicus- Role of hemodynamic shear stress in cardiovascular disease. Atherosclero- pid aortic valve disease: flow pattern is a major contributor and differs sis. 2011;214:249–256. doi: 10.1016/j.atherosclerosis.2010.09.008 with valve fusion type. Circ Cardiovasc Imaging. 2013;6:499–507. doi: 29. Ku DN, Giddens DP, Zarins CK, Glagov S. Pulsatile flow and atherosclerosis 10.1161/CIRCIMAGING.113.000528 in the human carotid bifurcation. Positive correlation between plaque loca- 11. Bieging ET, Frydrychowicz A, Wentland A, Landgraf BR, Johnson KM, tion and low oscillating shear stress. Arteriosclerosis. 1985;5:293–302. doi: Wieben O, François CJ. In vivo three-dimensional MR wall shear stress esti- 10.1161/01.atv.5.3.293 mation in ascending aortic dilatation. J Magn Reson Imaging. 2011;33:589– 30. van Ooij P, Potters WV, Collins J, Carr M, Carr J, Malaisrie SC, Fedak PW, 597. doi: 10.1002/jmri.22485 McCarthy PM, Markl M, Barker AJ. Characterization of abnormal wall shear Arterioscler Thromb Vasc Biol. 2020;40:e10–e20. DOI: 10.1161/ATVBAHA.119.313636 January 2020   e19 Dux-Santoy et al Oscillating Flow and Bicuspid Aortopathy stress using 4D flow MRI in human bicuspid aortopathy. Ann Biomed Eng. al. Decreased rotational flow and circumferential wall shear stress as early CLINICAL AND POPULATION 2015;43:1385–1397. doi: 10.1007/s10439-014-1092-7 markers of descending aorta dilation in Marfan syndrome: a 4D flow CMR 31. Vlachopoulos C, O’Rourke MF, Nichols WW. McDonald’s Blood Flow in Arter- study. J Cardiovasc Magn Reson. 2019;21:63. doi: 10.1186/s12968- STUDIES - AL ies: Theoretical, Experimental and Clinical Principles. 6th ed. London: CRC 019-0572-1 Press; 2011. doi:10.1201/b13568 35. Kontopodis N, Tzirakis K, Ioannou CV. The obsolete maximum diameter cri- 32. Kilner PJ, Yang GZ, Mohiaddin RH, Firmin DN, Longmore DB. Helical and terion, the evident role of biomechanical (Pressure) indices, the new role of retrograde secondary flow patterns in the aortic arch studied by three- hemodynamic (Flow) indices, and the multi-modal approach to the rupture directional magnetic resonance velocity mapping. Circulation. 1993;88(5 pt risk assessment of abdominal aortic aneurysms. Ann Vasc Dis. 2018;11:78– 1):2235–2247. doi: 10.1161/01.cir.88.5.2235 83. doi: 10.3400/avd.ra.17-00115 33. Richter Y, Edelman ER. Cardiology is flow. Circulation. 2006;113:2679– 36. Barker AJ, Markl M, Bürk J, Lorenz R, Bock J, Bauer S, Schulz-Menger J, 2682. doi: 10.1161/CIRCULATIONAHA.106.632687 von Knobelsdorff-Brenkenhoff F. Bicuspid aortic valve is associated with 34. Guala A, Teixido-Tura G, Dux-Santoy L, Granato C, Ruiz-Muñoz A, altered wall shear stress in the ascending aorta. Circ Cardiovasc Imaging. Valente F, Galian-Gay L, Gutiérrez L, González-Alujas T, Johnson KM, et 2012;5:457–466. doi: 10.1161/CIRCIMAGING.112.973370 Downloaded from http://ahajournals.org by on March 17, 2022 e20   January 2020 Arterioscler Thromb Vasc Biol. 2020;40:e10–e20. DOI: 10.1161/ATVBAHA.119.313636