Emerging imaging and circulating biomarkers in relation to underlying mechanisms in Bicuspid Aortic Valve aortopathy.

Previous studies have assessed the utilisation of various imaging modalities and potential imaging biomarkers that can be derived from them. One important study by Guzzardi and colleagues analysed this further by assessing WSS via 4D-flow CMR in BAV and healthy control patients and then further analysed the histology from aortic tissue samples in the BAV patient group [29] . 20 BAV patients undergoing ascending aorta resection were assessed and had WSS regionally mapped. Paired samples from the aortic wall were taken, correlating to areas of increased and normal WSS [29] . Segments of the aortic wall exposed to elevated WSS showed a decreased quantity of elastin (measured through Verhoeff-Van Gieson histological staining) compared to aortic wall that was exposed to normal WSS within the same individual. At a cohort level, elastin was significantly decreased in areas of elevated WSS in the BAV population, 36.61 ± 16.87 % vs. 49.12 ± 16.53 %; p = 0.04 [29] . Areas of elevated WSS had increased levels of Transforming Growth Factor Beta 1 (TGFβ-1) protein compared to regions of normal WSS in the same aorta (29). Elevated WSS area also demonstrated increased concentrations of Tissue Inhibitor of metalloproteinases- 1 (TIMP-1) p = 0.04, Matrix Metalloproteinase- 1 (MMP-1) p = 0.03, MMP-3 (p = 0.02) [29] , suggesting enhanced tissue remodelling. The study observed different patterns of dilation in the ascending aorta were associated with location of BAV cusp fusion. This highlights another factor that must be considered when assessing biomarkers for aortopathy. Nevertheless, identifying concentrations of MMP, TIMP, elastin and TGFβ-1 protein in areas of different WSS is a vital process in potentially understanding and also gaining new perspective into the complex cause of aortopathy in the BAV patient group and further supports the role of abnormal haemodynamics as an important factor that must be considered when prognosticating risk.

AS is known to be an independent risk factor for the development of aneurysm where previous studies have shown abnormal blood flow in those with TAV to develop thoracic aortic dilatation [52] . Yet it has been shown even when AS or AR are absent, BAV is still associated with increased WSS [52] . AS and AR may develop in the later stages of life in those with BAV, however the specificity of specific flow profiles dependent on valve anatomy still shows potential utilisation of this imaging biomarker in the earlier stages of the condition to possibly aid in the risk stratification of those more likely to develop aortopathy [16] , [53] . With reverse flow, even when AS or AR are absent, BAV can cause jet flow patterns that voxel-wise reverse flow studies can capture. This is essentially an imaging technique that can measure local abnormal flow specific to those with BAV, showing the specificity of reverse flow [20] . The assessments of the association of miRNAs with aortopathy are still at an early stage, in that research is currently in the identification phase of specific miRNAs associated with specific valvular profiles. This is a promising field of research as studies begin to show miRNAs associated with BAV and separate miRNAs associated with AS present in TAV patients [34] . The preliminary stage of research can also be extended to MMPs, as research suggests their involvement (through increased expression) in the reduced distension and increased dilatation of the aortic wall in those with BAV [54] . There are 23 MMPs known to be expressed in human tissue, the abnormal WSS is thought to act on the endothelial surface and possibly lead to the induction of cellular signalling cascades which ultimately leads to increased expression of certain MMPs such as MMP-2 and MMP-9 [54] . These MMPs can be measured from the plasma, it is known BAV with AS can create a different WSS pattern from studies previously discussed and raises the question, does this distinct flow pattern ultimately result in the expression of different MMPs at the cellular level which can measured from plasma. An interesting study by Wang and colleagues, observed such a difference where plasma MMP-9 was only increased in isolated severe stenotic BAV patients [55] . Yet in echocardiographically normal BAV subjects, MMP-2 was the only independent predictor for increased aortic diameter [55] . Raised uDES levels has shown an association with aortic diameter in those with BAV with another study showing the association between pDES and those with AAA [47] , [48] . The analysis of aortic tissue from those with BAV has clearly demonstrated loss of elastin at points subjected to increased WSS, irrespective of valvular lesions [29] . Further work would be needed to directly assess if a worsening degree of AS or AR would affect pDES levels; this would be assuming that the AS or AR contributed to the dilatation of the aorta. The readily availability of troponin T and NT-proBNP plasma blood tests would make them an ideal candidate for use as clinical biomarkers, however, their specificity for BAV associated aortopathy has not been shown in previous studies [7] . The levels of both can be raised but in the context of subjects developing AS or AR where increased levels are due to ventricular dysfunction caused by this valvular pathology [7] . Table 3 summarises the potential clinical utilisation and current challenges of the biomarkers discussed. Table 3 Utility of biomarkers in BAV associated aortopathy and current challenges. Biomarker Possible Utilisation Current Challenges References CT Assessment of Tortuosity Current CT technology can be used for measurement and assessment of aortic tortuosity allowing identification of aortas vulnerable to dilatation. Where CT scanning is already used for assessment of aortic diameters, additional measurement could enhance the risk stratification for dissection risk. Tortuosity could be underestimated due to the 2D imaging techniques. Limited data on aortic tortuosity progression over time is available. The risk of radiation exposure from extended and frequent CT scanning needs to be considered when techniques such as 4D Flow CMR are becoming more accessible. More evidence is needed to initially establish the relationship between tortuosity and aortopathy in BAV. [63] 4D Flow CMR Wall Shear Stress Assessing areas of the aorta subjected to increased abnormal blood flow resulting in tissue remodelling. Clinicians could potentially identify patients at increased risk of aortopathy and rupture. 4D flow CMR can be used to establish a connection between WSS and biomarkers for inflammation and collagen synthesis. 4D Flow CMR is only now beginning to be used routinely in clinical use, but quality assurances on different post-processing platforms available remains to be established. Standardised and validated imaging processing software is required. [17] 4D Flow CMR Reverse Flow Could be utilised to assess abnormal aortic blood flow even in the absence of AS or AR in BAV patients. More elevated flow reversal in BAV is associated with more severe aortopathy and could potentially help determining timing of surgery in addition to current 2D size parameters. However, this potential use as a prognosticator requires further study. As previously mentioned, there is a need for automated processes when utilising 4D Flow CMR for standardisation and to reduce the high operator labour intensity. In addition, there is an unmet need to transfer research scanning techniques to clinical use. [17] MMP Plasma MMP levels can be measured showing utility in assessing ongoing disease activity of aortopathy for disease prognosis in standard or outpatient setting. Prognostic role is not yet determined. Further studies are still needed to assess the association between MMPs such as MMP-2 with the progression of BAV induced aortopathy. [55] miRNA Can be measured in plasma and urine to determine ongoing disease activity of aortopathy in the clinical environment. Specific miRNAs beginning to show association to BAV and aortopathy, but prognostic value remains to be determined. Currently, further research is needed in identifying which miRNAs are associated with BAV induced aortopathy and if they can be differentiated from valvular lesions such as AS or AR. Few studies in general have advanced to the clinical trials stage when assessing the clinical applications of miRNAs in various medical context. However, there is evident progression of understanding their role in disease pathophysiology and potential use as treatment. [64] Troponin and BNP Can be measured in plasma and urine in to determine the effect of BAV to the heart in the clinical environment. Increased plasma levels could signify worse cardiac burden consequent to a higher grade of valve pathology which could contribute to aortopathy development however a direct link to aortopathy for both has not been observed. Troponin T is specific for myocardial cell necrosis and BNP in ventricular walls strain, but neither is directly increased in the plasma in response to vascular remodelling which occurs in aortopathy. Therefore, little evidence currently supports their use in patients with BAV with no valvular lesions. [7] TGFβ1 Current evidence does not support its clinical utilities in assessing BAV associated aortopathy. Elevated total TGF-β1 levels have been observed in the whole spectrum of genetic aortic syndromes. There has been significantly different levels between those with aortic dilatation or aortic valvulopathy and the healthy population but differences depending on valve type i.e. BAV or TAV is less apparent in these studies. [32] , [38] , [44] Desmosine Can be measured in plasma and urine in to determine ongoing disease activity of aortopathy in the clinical environment. Combining imaging and plasma biomarker modalities could be more informative in assessing the risk of aortic dissection. The increases of desmosine concentrations in plasma and urine lack disease specificity. Literature has suggested variable pDES levels in those with lung pathology such as COPD, this could create challenges in interpretation of results in a BAV patient for instance who developed COPD. [46] , [47] , [49] , [62] BAV, Bicuspid Aortic Valve; MMP, matrix metalloproteinase; TIMP, tissue inhibitors of metalloproteinase; TAV, tricuspid aortic valve; TGFβ1, Transforming growth factor beta 1; miRNA, MicroRNAs; CT, Computed Tomography; MRI, Magnetic resonance imaging; AS, aortic stenosis; AR, aortic regurgitation; AV aortic valve; TTE, transthoracic echocardiogram; WSS, wall shear stress; NT-proBNP, N-terminal pro b-type natriuretic peptide; hsTnT, high-sensitive troponin T; COPD, Chronic Obstructive Pulmonary Disease; WSS, Wall Shear Stress; 4D-CMR, 4-dimensional cardiovascular magnetic resonance imaging. Utility of biomarkers in BAV associated aortopathy and current challenges. BAV, Bicuspid Aortic Valve; MMP, matrix metalloproteinase; TIMP, tissue inhibitors of metalloproteinase; TAV, tricuspid aortic valve; TGFβ1, Transforming growth factor beta 1; miRNA, MicroRNAs; CT, Computed Tomography; MRI, Magnetic resonance imaging; AS, aortic stenosis; AR, aortic regurgitation; AV aortic valve; TTE, transthoracic echocardiogram; WSS, wall shear stress; NT-proBNP, N-terminal pro b-type natriuretic peptide; hsTnT, high-sensitive troponin T; COPD, Chronic Obstructive Pulmonary Disease; WSS, Wall Shear Stress; 4D-CMR, 4-dimensional cardiovascular magnetic resonance imaging.

Hamza M Ahmad: Writing – review & editing, Writing – original draft, Methodology, Conceptualization. Zaid Iskandar: Writing – review & editing, Conceptualization. Chim C. Lang: Writing – review & editing, Writing – original draft. Jeffrey T.J. Huang: Writing – review & editing, Writing – original draft, Conceptualization. Anna-Maria Choy: Writing – review & editing, Writing – original draft, Supervision, Conceptualization.

With risk assessments currently based on aortic diameter still being used in the BAV aortopathy population, further work is needed in not only identifying biomarkers but also evaluating their clinical utility. 4D-Flow CMR has created an opportunity to identify/calculate different imaging parameters which could be used to find novel imaging biomarkers such as WSS. CT imaging is limited in value when calculating blood flow velocities with limited planes in comparison to 4D imaging [17] . Further work is needed in assessing these image parameters in subjects where there is adequate regular follow-up of aortic growth to truly assess correlations. Prospective studies with larger patient cohorts are needed when reviewing candidate circulating biomarkers. Another challenge is that BAV subjects in some studies already have a confirmed aneurysm/aortic dilatation, ideally studies with a BAV population with normal aortic diameters would be needed to assess the true effectiveness of risk stratification with selected biomarkers. Nevertheless, the focus on identifying relevant biomarkers has led to deeper understanding of mechanisms involved in the pathogenesis of aortopathy.

The challenge in confirming an estimate of the incidence of aortopathy in the BAV population is due to its asymptomatic course until an adverse event occurs hence cases are often observed incidentally or via routine surveillance imaging. The mechanism of how aortopathy occurs is still uncertain, initially thought to be genetically driven, however development of new imaging such as cardiac magnetic resonance 4D- Flow imaging (4D-flow CMR) has illustrated the role of haemodynamic abnormalities contributing to pathology [16] . This has led to the recognition of flow-related measurements as potential prognosticators of acute aortic syndromes occurring in the BAV patient group.

CMR is the gold standard imaging practice for the assessment of cardiac function, the advantage with this imaging modality is its flow detectability and volume analysis which can create a 3 dimensional map of flow patterns, jet streams, velocity and volume of blood at specific segments of the aorta. This allows for new perspectives in understanding the pathogenesis of vessel dilation [16] . TTE is limited in its reproducibility due to interobserver variation, Computed Tomography (CT) is able to provide 3D scans that can demonstrate the spread of contrast through blood vessels in that moment of time. 4D Flow imaging can demonstrate blood flow and variable haemodynamics through 3 spatial directions with incorporation of time, the fourth dimension [17] . Velocity vectors of blood flow gained from these images can be utilised to analyse blood flow and various parameters related to this. These are vital during an era of research where more evidence illustrates the potential relationship between abnormal haemodynamics and aortic pathology such as the development of aortic aortopathy [18] .

Two such measurements are reverse flow (RF) and forward flow (FF) derived from 4D imaging, a previous study by Geeraert and colleagues reviewed these components in individuals with BAV and the associations with aortic dilatation [19] . 96 subjects, 73 BAV and 23 health controls, underwent cardiac 4D-Flow CMR imaging with 4D flow-analysis carried out in 5 regions of the aorta based on previous thoracic aortic disease guidelines with calculation of maximum aortic diameter. The BAV cohort demonstrated greater helical flow areas with elevated RF throughout the aorta, the main increase in flow at the ascending aorta, 254 %, p < 0. 001 [19] . Patients in this group with moderate dilatation of the aorta showed increased RF in the arch, 83 %, p < 0.05, compared to non-dilated subjects. BAV subjects with severe dilatation showed even more RF at the arch, 106 %, p < 0.01 [19] . This elevated RF is possibly due to the significant de-centred jet produced by the bicuspid valve itself, even further different flow patterns were observed depending on valve phenotype [19] . Following this a retrospective cross-sectional study by Weiss and colleagues further looked at relationship between reverse flow and the extent of aortic dilation in BAV patients [20] . There were 510 BAV subjects and 120 with a trileaflet aortic valve (TAV) with 25 healthy controls. Again, 4D-Flow CMR was utilised with an in-house programme to calculate parameters such as FF, RF, automated aorta centre line for jet analysis [20] . In the BAV group with isolated AS and no AR compared to healthy control subjects, RF was increased in the ascending aorta in the BAV group. To note, the more severe the degree of AS there was a marked increase in RF, this was noted in systole. Diastolic reverse flow in the ascending aorta was not dissimilar depending on degree of AS in the BAV group [20] . Reverse flow in the ascending aorta in BAV patients without AS or AR was increased compared to the controls- 94 % increase, p < 0.001. Further to this, systolic flow was increased in BAV patients without AS or AR compared to TAV patients who also had dilatation (79.3 % increase, p < 0.0001) showing specificity to BAV patients [20] . As previously, depending on the BAV valve morphology there was discrepancy in the systolic reverse flow. Increase systolic and diastolic reverse flow was associated with a larger mid-ascending aorta diameter, R = 0.22, p < 0.0001 and R = 0.09, p < 0.05 respectively. Voxel-wise reverse flow maps derived from 4D flow CMR utilising in-house programmes were used to observe the flow disruptions caused by BAV in systole and diastole, even when AS and AR is absent [20] . Compared to the previous study which included significant manual interaction with the analysis and use of an in-house programme, the study by Weiss and colleagues utilised a ‘semi-automated method’ with re-assuring interobserver agreement [20] . The study highlighted a degree of variability when processing these CMR scans and a potential area for further work to achieve standardisation in the clinical setting. Studies have shown an association between RF and aortic diameter, further research into how reverse flow correlates with aortopathy progression is needed to further assess its utility as an imaging biomarker. Wall Shear Stress (WSS) is another potential biomarker for aortopathy in the BAV cohort defined as the frictional forces produced by the flow of blood at the wall of a blood vessel. The measurement, calculated from 4D-Flow CMR imaging has been shown to be abnormal in previous studies and can be expressed in different forms such as net magnitude WSS, and its vector components circumferential WSS/ axial WSS as shown in Fig. 2 [21] , [22] . Fig. 2 WSS and Components. Fig. 2 demonstrates image of 4D Flow CMR and graphical overview used to determine WSS and its various parameters such as WSS angle, magnitude, axial and circumferential. The association with this and aortic aneurysm formation can be viewed (Minderhoud et al. Adapted from [22] . WSS and Components. Fig. 2 demonstrates image of 4D Flow CMR and graphical overview used to determine WSS and its various parameters such as WSS angle, magnitude, axial and circumferential. The association with this and aortic aneurysm formation can be viewed (Minderhoud et al. One such study by Guala and colleagues assessed the predictive value of WSS on ascending aorta growth rate in the BAV cohort [21] . 47 patients with BAV were analysed, they had underwent 4D-Flow CMR and 2 computed tomography angiography scans (at least 6 months apart), enabling follow-up of growth rate of aortic diameter. WSS and its circumferential component showed statistically significant associations with the aorta growth rate at the level of the pulmonary artery bifurcation (R = 0.291, p = 0.049 and R = 0.0358, p = 0.014 respectively). Circumferential WSS was associated positively with areas of the aorta that had the fastest growth during follow-up [21] . Minderhoud et al investigated the association between WSS measurements and volumetric growth in the ascending aorta using longitudinal data in the BAV cohort where 28 age-matched healthy controls were utilised, both groups had baseline 4D flow CMR carried out and the 32 BAV patients further had a baseline CTA and a follow-up CTA 3 years later. There was no statistical significance of WSS magnitude between both groups at the proximal ascending aorta. Circumferential WSS and WSS angle were higher in the BAV group (p < 0.001). WSS angle and volumetric were significantly related, r = 0.41, p = 0.02), the entire ascending aorta was noted to have trends towards more volume growth with increasing circumferential WSS and WSS angles, although, these weren’t significant, r = 0.45, p = 0.069 and r = 0.45, p = 0.071 respectively. WSS angle was the only parameter to be shown to be associated with aortic growth ( β  = 0.121, p = 0.031), which remained significant after adjusting for diastolic blood pressure and aortic volume emphasising use in helping to identify patients at risk of aortic growth [23] . It is not clear why WSS magnitude was not elevated in the BAV patient group, WSS is known to lessen as a blood vessel dilates and with BAV patients in general being more susceptible to larger aortic diameters this could potentially explain the lack of difference between both groups [23] . This also highlights further research with larger patient groups investigating WSS and its use as a predictor of aortic growth, its components such as angle, axial and magnitude must be considered with the noted variation in previous studies but they do highlight altered haemodynamics and more importantly specificity to the BAV cohort which could be utilised in the clinical setting. A systematic review, focusing on the use of WSS in risk stratification of BAV aortopathy reviewed 26 studies that utilised 4D flow CMR to analyse WSS in the ascending aorta. Several studies showed an association with aortic jet flow patterns and WSS, helical flow pattern seen in the BAV cohort as compared to the healthy TAV cohort was associated also with larger ascending aorta diameter and elevated magnitude WSS and circumferential WSS [16] . When specifically looking at WSS the effect of AS must also be taken into consideration with several studies also adjusting for this variable in their models [16] , [24] , [25] . Furthermore, Shan and colleagues investigated the effects of severe aortic stenosis on WSS and flow patterns in BAV patients with a specific phenotype [24] . 4D-Flow CMR was utilised in 120 BAV subjects and 20 TAV controls. Altered outflow patterns from the aorta in the BAV group were observed at in either the right-anterior or right-posterior regions of the ascending aorta depending on valve morphology, severe AS was noted to cause a further acceleration in these outflow jets. Differences in vortical, helical and peak systolic flow were apparent depending on BAV morphology when compared to the TAV group. However, importantly these differences in peak velocity distributions between subsets of the BAV group categorised based on valve type weren’t evident at the level of mid-ascending aorta if the patient had severe AS. Differences in one of the components of WSS in the aortopathy group, again were undetectable when severe AS was present and it contributed to patient to patient variability in WSS assessment [24] . This creates another factor that must be considered when reviewing WSS as a potential imaging biomarker in that reproducibility for identifying those who may develop aortopathy may not be consistent, defined thresholds incorporating the effect of AS may need to be created when assessing its use in clinics. Arterial tortuosity has been shown to be a predictor of aortic events in heritable aortopathies [26] . This morphology could possibly reflect remodelling in a pathological process due to worsening structure of the aorta in BAV patients secondary to genetic mutations in the BAV population or the abnormal haemodynamic flow that has been observed on imaging [27] . One study investigated this further using a 2D- CT scan method to analyse aortic tortuosity in a BAV cohort to analyse its usefulness in identifying patients at risk of developing aortic complications [27] . To calculate this measurement, a triangle was drawn between arch apex and midline of the thoracic ascending aorta and thoracic descending aorta at the pulmonary artery level. Arch length was the sum of these segments, arch angle was the vertex angle in the triangle drawn and arch width was the length of the base of the triangle. Tortuosity itself was calculated as the arch length divided by arch width. 354 subjects were analysed in the study, 120 BAV patients and 234 TAV patients matched by age and gender. Aortic arch tortuosity was significantly higher in the BAV group compared to the TAV group. (median 1.76; interquartile range [Q1-Q3: 1.62 − 1.95] vs. 1.63 [1.53 − 1.78], p < 0.01). When using multi-variable analysis adjusting for maximum aortic diameter, race, smoking this association was still independently associated [27] . When utilising CT imaging for arch tortuosity, the degree of this measurement may be underestimated by using 2D imaging measurements- arch lengths could extend in more than one plane; thus highlighting a limiting factor. Further work could potentially involve focusing on tortuosity progression and the relationship to the aortic diameter as time progresses to explore it’s usefulness [28] . Another study by Yeats and colleagues observed the 3D characterisation of the aortic arch in patients with BAV who were pre-transcatheter aortic valve replacement (TAVR) with aortic stenosis through CTA [28] . 59 patients with AS and BAV who were to undergo TAVR had images analysed with 3D geometric parameters to calculate maximum curvature in the aorta as well as assess BAV morphology- the study outlined how areas of increased curvature within the aorta are known to disrupt flow within the vessel. Yet it is still unknown how these high curvature areas and diameter effect progression of disease within the aorta. When re-constructing the images, segmentation was used in this study, the operator variability was assessed by having 2 operators participate in the study when segmenting [28] . The study mentions how operator variability was not significant but raises the question of its increased impact in hypothetical clinical settings and the need for acceptance criteria when assessing thresholds for various measurements at different locations within the aorta [28] . From studies such as this, it is evident there is change to the structure of the aorta, whether tortuosity or curvature changes can be used to monitor aortic dilation could be an area for potential work.

Around 50 % of patients with BAV will develop aortopathy, with more than 25 % requiring surgical intervention. There is a recognised clinical need to develop better methods for monitoring the risk of aortic dissection in this patient group. This review highlights recent developments in imaging and peripheral biomarkers. Although no biomarkers have been fully validated, 4D-flow CMR-based imaging, miRNA-17, and desmosine show promising potential. Future studies should focus on standardising methods (e.g. imaging processing) and exploring the longitudinal associations between biomarkers, disease progression, and how these biomarkers can be integrated into therapeutic interventions.

Transforming Growth Factor Beta 1 (TGFβ-1) has been shown to play a role in aneurysm formation in the thoracic and abdominal aorta [30] . A previous study further showed elevated total serum TGFβ-1 in patients with BAV, Marfan Syndrome, Loeys-Dietz Syndrome (LDS) and thoracic aortic aneurysm [30] . Elevated serum levels were observed in patients with BAV and MFS compared to those with a genetic aortic syndrome [25] . The mechanism underlying aortopathy is yet to be fully established, TGF-β1 dysregulation is thought to contribute to impaired vascular matrix homeostasis. TGF-β1 itself is a cytokine involved in other various pathways associated with cell proliferation, extracellular matrix (ECM) remodelling as well as aortopathy [31] , [32] . One study assessed, 50 patients who had either TAV or BAV and normal or dilated aorta. 30 samples from the aorta were assessed by a pathologist, 40 control patients were involved in the study to establish ranges for TGF-β1. TGF-β1 concentrations demonstrated a significant difference between the BAV and control group (p < 0.001) where concentrations were reduced in the BAV group but no statistical difference in levels when patients were categorised by aortic valve phenotype, i.e., TAV vs BAV [32] . Further studies have also assessed plasma levels of TGF-β1 to monitor progression of disease such as psoriasis to multiple myeloma [32] . The extensive role this marker plays in many disease states limits its use for risk stratification in the BAV cohort as it’s specificity must be questioned when it is involved in multiple pathways. MicroRNAs (MiRNA) have also been explored as potential biomarkers for BAV aortopathy, these are single stranded RNA that control gene expression through binding to messenger RNA (mRNA). The interest and advantage in utilisation of MiRNAs as a biomarker is their resistance to degradation with one study demonstrating no significant difference in miRNA serum levels that were stored for 10 days compared to storage for 20 months and up to 2 years [33] . A study by Girdauskas and et al assessed miRNAs in 145 individuals referred for aortic valve surgery to assess correlation between miRNA and BAV aortopathy [34] . 63 patients had bicuspid valve AR, 32 with bicuspid valve AS and 50 had TAV. Aortic diameters were measured via TTE at the Sinus of Valsalva. miRNA was obtained from serum samples, the study selected miRNAs based on previous published literature which associated specific miRNA with BAV/TAV associated related aortopathy and aortic aneurysm formation [34] . Correlation analysis between maximum aortic diameter and miRNA showed a significant inverse linear correlation with miR-17 and miR-20a, r = -0.285: p = 0.005 and r = -0.215: p = 0.035 respectively. In the BAV group with AS specifically there was a significant inverse correlation between values of miR-17 and miR-20a ( r  =  − 0.479, p  = 0.01 and r  =  − 0.378,  p  = 0.045) respectively. The mechanism through which miRNA-17 is thought to affect aortopathy is via regulation of TIMP expression [34] . Another study aimed to assess miRNAs ability to predict BAV aortopathy in 63 BAV patients who had undergone aortic valve with or without proximal aortic surgery [35] . Follow-up was a mean of 9.3 ± 5.7  years which included aortic imaging and blood sampling. Adverse events in the study consisted of re-doing aortic surgery or increasing aortic root diameter. Patients where split into those with a severely dilated aortic root (>/=50 mm) and less dilated aortic root (<50 mm). miRNAs in the severely dilated group were generally lower, and miR-17 and 106a were significantly increased in the group with lower aortic diameters. And again miR-17 and miR-106a were significantly lower in patients who had progression of aortopathy which was defined as a diameter increase of >/= 3 mm. Those who faced adverse aortic events had significantly downregulated miR-17 and miR-106a, delta Ct (cycle threshold) 1.51 ± 0.73 vs delta Ct 2.00 ± 0.61, P = 0.02 and delta Ct 5.39 ± 0.69 vs delta Ct 5.85 ± 0.44,  P  = 0.007 respectively [35] . miRNA selection for analysis, again like the previous study utilised reports of miRNA associated with aortopathy from previous literature. Various miRNA gene clusters have been shown to inhibit TIMP expression which increases MMP-2 activity contributing to ECM breakdown and development of aortopathy [36] . Despite the promising data from miR17/106a, the specificity to aortopathy needs to be studied further. For instance, a prospective laboratory study in China found miR-17 to be downregulated in women with endometriosis compared to those without, highlighting the need for further research [37] . Matrix Metalloproteinases (MMPs) are enzymes that participate in the degradation of components of the ECM. Their increased activity has been identified in the BAV cohort in comparison to those with TAV, MMP- 2 has been identified as a potential biomarker with previous studies showing association with the proximal aortic diameter [38] . One such study by Wilton and colleagues investigated the difference between the ascending aorta matrix components in BAV patients vs TAV. 84 patients were selected who were undergoing surgery for the AV or aorta, 54 patients had TAV and 28 had BAV. A subset of 19 patients with a dilated ascending aorta, diameter >/=4.0 cm, were analysed, there was a greater ratio of MMP-2/TIMP-1 gene expression in the ascending aorta of the patients with BAV compared to TAV, p = 0.05. TIMP and MMP balance in tissue is thought to regulate the degradation of the ECM in normal and diseased states with a potential mechanism of abnormal haemodynamic flow affecting regional transcription/translation of these MMP and TIMPs leading to ECM breakdown [39] . Ikonomidis and colleagues hypothesised that mechanisms involved with aneurysm formation in the ascending thoracic aorta associated with BAV would be different compared to those with thoracic aortic aneurysms and TAV with a focus on MMP and TIMP expression. 53 patients with BAV and 46 with AV had aortic specimens analysed, (taken during aneurysm resection/AVR). This study used a control group of 26 participants where ascending aortic specimens were taken from patients who underwent: coronary bypass, were heart transplant recipients, heart/lung donors with none in this group having dissection. MMP-2 was increased in the BAV group compared to the TAV and control group. The study demonstrated significant correlations between MMP-14, MMP-2 and aneurysm size (rho = -0.313, p = 0.003, rho = 0.243, p = 0.02 respectively) [40] . In the TAV group, correlations between MMP-7, TIMP-2 and aneurysm size were observed, where MMP-7 was increased in subjects with larger aneurysm size compared to smaller size groups [40] . The findings of this study are interesting in that previous to this, a comparative study of comprehensive analysis of the MMP and TIMP profiles that occur within aortic aneurysms of those with BAV and TAV had not been carried out. From the study it can be seen that specific profiles exist depending on valve type potentially showing utilisation of specific MMPs as biomarkers for aneurysm size [40] . Troponins are proteins located in the troponin complex in cardiac and skeletal muscle. The complex contains 3 subunits: I, T and C. In combination with calcium ions, troponins have a regulatory role in muscle contraction [41] . Troponin C synthesis occurs in skeletal and cardiac tissue whereas T and I are mainly specific to the myocardium. Patients with BAV tend to have an increased degree of stenosis or regurgitation which can lead to elevated levels of troponin-T [7] . Brain natriuretic peptide (BNP) is a 32 amino acid peptide, predominantly found in the ventricles, produced from its precursor- proBNP [42] . Both active BNP hormone and its inactive N-terminal fragment (NT-proBNP) are measurable in serum. Various conditions can lead to secretion of BNP such as those causing increased myocardial stretch with high pressure filling. The action of BNP is to counteract the pathophysiological process leading to progression of heart failure via inducing natriuresis and diuresis [42] . A prospective study aimed to assess the value of troponin T and NT-proBNP as biomarkers in patients with BAV. Data utilised were from two different cohorts of patients with BAV used in previous studies who also underwent TTE [43] . 182 patients in total were included, 23 % with left atrial enlargement, 35.2 % with severe aortic stenosis and 43.4 % with regurgitation. Troponin T was found to be higher in males compared to females. NT-proBNP was lower in men compared to females, median follow-up was 6.9 years. Elevated NT-proBNP demonstrated strong associations with the composite endpoints of arrhythmias or death [43] . This could represent the increased work of the myocardium during the early phase of BAV. The prognostic value could be extended to the later development of aortopathy in the BAV group. The current study demonstrating the association of increased levels with severe endpoints and the assumption that aortopathy in BAV patients develops as the effects of the condition become more severe. The sex differences between males and female in the study would have to be considered when relating this to clinical use. Troponin T did not show a significant association with the endpoints [43] . Bons and colleagues in a cross-sectional study assessed the association in biomarkers including high sensitivity troponin T and NT-proBNP with extent of AV stenosis/regurgitation and aortic diameter [44] . Patients utilised were from the data of 2 groups used in previous studies who had underwent TTE also which totalled 183 subjects. 85 healthy adults were recruited as age matched controls who also had TGFβ-1 measured. No biomarker in this study showed an association with aortic diameter, a two-fold higher troponin T was associated with an elevated risk of one higher grade of aortic regurgitation (factor of 1.34 95 % CI 1.01;1.76) [44] . Higher NT-proBNP was significantly associated with increased severity of aortic regurgitation and maximum velocity across the aortic valve. Although no association with aortic diameter, the readily availability of troponin-T and NT-proBNP in comparison imaging biomarkers in the clinical environment may be useful in identifying those with increased disease burden in the BAV population [44] . Elastin, a protein found in connective tissue that contributes to the elasticity of vasculature, is one of the main components involved in maintaining structure of the aorta [45] . The elastin network is formed by the crosslinking of tropoelastin monomers mainly synthesised from fibroblasts and vascular smooth muscle cells. The cross-links within elastin can be liberated to two amino acid isoforms: desmosine and Isodesomosine, which can serve as a biomarker of mature elastin breakdown which occurs in BAV aortopathy ( Fig. 3 , [45] , [46] ). Desmosine, upon elastin degradation, is released in a peptide form to circulation, sequestered by the kidney, and released almost exclusively to urine [46] . Circulating and urinary desmosine are currently being considered as a potential biomarker for BAV associated aortopathy. Fig. 3 Loss of Elastin. An example of an aortic dissection in a male patient with BAV. A) Aortic wall dissection involving intimal tear can be seen at the tubular ascending aorta. B) Weigert-Van Geison Stain demonstrating the tunica media with elastic fibre loss ( Adapted from Thiene et al. [10] . Loss of Elastin. An example of an aortic dissection in a male patient with BAV. A) Aortic wall dissection involving intimal tear can be seen at the tubular ascending aorta. B) Weigert-Van Geison Stain demonstrating the tunica media with elastic fibre loss ( A study by Iskandar and colleagues attempted to assess whether BAV patients had an increased elastin degradation as measured through urinary and plasma desmosine and to investigate the relationship between urinary and plasma desmosine levels [47] . Aortic root size and z-scores were measured via TTE in both BAV patients and healthy controls. Plasma desmosine (pDES) and urinary desmosine (uDES) were elevated in BAV patients when compared to controls (0.30 ± 0.10 vs 0.26 ± 0.075 ng/mL, p = 0.01 and 15.9 ± 4.6 vs 7.2 ± 2.8 ng/mg creatinine, p < 0.001 respectively) [47] . Further to this, uDES had a significant association with maximal aortic root size in comparison to controls; the study noted a significant correlation between urinary and plasma desmosine [47] . These results suggest urinary or plasma desmosine may have a potential utility in monitoring BAV associated aortopathy. The relevance of desmosine in other aortopathies has been shown in abdominal aortic aneurysms with 507 patients with AAA and 162 control subjects from 2 observational cohort studies [48] . In the longitudinal study, pDES was elevated in patients with AAA compared to controls, analysis also showed when pDES concentrations were stratified by aortic diameter there was an incremental stepwise increase with larger aneurysms being associated with greater pDES concentration [48] . This study strengthens the idea that circulating desmosine is more likely to reflect elastin degradation in aorta tissues. Abnormal elastin breakdown also occurs in other conditions such as chronic obstructive pulmonary disease (COPD) and bronchiectasis, which raises concerns regarding disease specificity. Interestingly the elevated circulating desmosine concentrations were predominantly associated with cardiovascular outcomes suggesting the elastin degradation may be more relevant to vasculature than the lungs [49] , [50] , [51] . Desmosine has shown potential as a biomarker that could be utilised in BAV associated aortopathy however further larger studies are needed to validate clinical use with a focus on specificity as a tool to help clinicians risk stratify in patients with BAV [20] .

Bicuspid Aortic Valve (BAV) is the most common congenital heart defect with a prevalence of up to 2 % and carries a 30 % lifetime risk of complications [1] . BAV is associated with genetic syndromes such as Turners syndrome, Loeys-Dietz syndrome and Marfan Syndrome, however, most cases are isolated and non-syndromic with no single gene variant identified and involve complex gene interaction [1] . One of the first descriptions of BAV was reported approximately 500 years previously by artist Leonardo da Vinci when he sketched this variant of the aortic valve [2] , [3] . Although BAV refers to the fusion of two aortic valve (AV) cusps, it is not a disease isolated to the aortic valve alone. There is heterogeneity in the valve phenotypes and clinical traits according to the specific valve variant with associated changes in cardiac and aorta development [1] , [4] . To address variety in nomenclature with previous literature, a recent International Consensus for the classification of BAV was published, identifying three main types of valve anatomy: the fused BAV, partial-fusion BAV and 2-sinus BAV. The fused BAV is the commonest with 90–95 % of cases presenting with this classification [4] , [5] . Bicuspid aortopathy refers to the dilatation of the aorta commonly seen in those with BAV, with ranges from 35-80 % from various reports [6] . Dilatation is typically observed in the ascending aorta, however, this can also occur in the aortic root, tubular ascending aorta and proximal aortic arch. Valve related disease, stenosis/insufficiency, aortic aneurysm and aortic dissection contribute to the increased mortality with BAV condition- more so than the combination of all other congenital heart lesions [6] , [7] . On average, 50 % of patients with BAV will develop aortopathy with more than 25 % requiring surgical intervention. Aortopathy itself is a risk factor for aortic dissection, which is associated with a higher mortality. BAV patients are at 8 times greater risk of developing this complication. Previous studies have shown this risk of dissection increases with aorta diameter size, however its role in predicting when dissection occurs has limitations [6] , [8] . Fig. 1 demonstrates a BAV associated aneurysm via TTE and CT imaging modalities [7] , [9] . Fig. 1 Displays of a BAV associated Aneurysm. A) TTE 2D imaging of the dilated ascending aorta, ventricles are in view also. B) Axial Plane CT showing the dilated ascending aorta. C) CT 3D reconstruction demonstrating an 80 mm aneurysm ( Adapted from Junco-Vicente et al. [7] ). Displays of a BAV associated Aneurysm. A) TTE 2D imaging of the dilated ascending aorta, ventricles are in view also. B) Axial Plane CT showing the dilated ascending aorta. C) CT 3D reconstruction demonstrating an 80 mm aneurysm ( The development of aortopathy in the BAV cohort is associated with risk of dissection and possible sudden death [10] . Thiene and colleagues reported on their 30 year experience of researching sudden death in those younger than 35 years old, aortic dissection caused death in 3.5 % of cases with half of the dissections attributed to BAV associated aortopathy [10] . Importantly, in two of the cases, rupture occurred despite maximal aortic diameter being below the surgical threshold for intervention [10] . The widening use of Transthoracic Echocardiogram (TTE) in the healthcare setting has created the opportunity for routine clinical screening of patients with BAV for assessment of cardiac function, aortic diameters and valvular pathology. Valvular pathologies such as Aortic Stenosis (AS) and Aortic Regurgitation (AR) have been studied extensively and their association with BAV is established; however, the development of complications in BAV patients without these valvular pathologies is less known [11] . One study assessed TTE scans of 1307 BAV patients over a 15 year period, including 187 (25 %) who had no aortic stenosis or aortic regurgitation and found aortopathy to be present at the initial TTE in 80 (43 %) [11] . At 6.0 ± 3.8 years post-diagnosis, 15 % developed severe aortopathy, with a statistically significant increased rate of aorta surgery in those with aortopathy compared to those without, 21 % vs 1.8 % p < 0.05 [11] . This signifies aortopathy can occur in BAV patients without severe valvular pathology and is associated with increased risk of aortic surgical intervention years later showing the risk aortopathy itself poses. Furthermore, a systematic review focused on factors contributing to the growth rate in the thoracic aorta, assessed 11 studies that utilised Computed Tomography (CT), Magnetic Resonance Imaging (MRI), Echocardiography or a CT/MRI combination to evaluate aneurysms. The majority of studies had a scanning interval of 6 months with a range 2 to 24 months [12] . Results demonstrated that compared to those with degenerative aortopathy, those with BAV have a significantly higher thoracic aortic growth rate (0.42 mm/year vs 0.2 mm/year p = 0.02). Additionally, two studies demonstrated a trend towards increased intervention rates in those with a higher aortic growth rate [12] . Aortic diameter has been utilised for the indication of aortic surgery. The recent 2021 European Society of Cardiology Guidelines advises, in patients with BAV, aortic diameter greater than 5.5 cm warrants surgical intervention [13] . If other risk factors are also present, such as moderate to severe aortic stenosis, aortic regurgitation, or the presence of a dilated aortic root sinus [14] , the threshold aortic diameter for intervention is greater than 5 cm [13] . More recently, it has been recognised there is a gap between image-related management and the true real risk of aortic events in the BAV cohort. Strong evidence from the International Registry of Acute Aortic Dissection shows that, among patients with Type A dissection, 59 % had an aortic diameter below 5.5 cm and 40 % below 5 cm. These studies recognise that imaging modalities, while useful in monitoring aortic growth and other parameters, do not always correlate with the actual risk of aortic events, such as aortic dissection or rupture, highlighting the need for a better valid biomarker in predicting these adverse outcomes [8] , [15] . A study by the Healthcare Safety Investigation Branch in the UK in 2020 reported 20 % of patients with acute aortic dissection die before reaching hospital and 50 % will die before reaching a specialist centre in England [9] . Given the limitations of relying on aortic measurements to identify those at risk, it is vital that further research is conducted to identify better biomarkers of aortopathy in the BAV population. This review will attempt to assess and summarise recent studies identifying potential imaging biomarkers ( Table 1 ) and circulating plasma biomarkers ( Table 2 ). Table 1 Previously studied imaging biomarkers utilising BAV subjects. Biomarker Study Sample Method Key Results References WSS 47 BAV patients Median Age 46.7 years Underwent 4D-CMR Flow imaging and baseline/follow-up CTA for Aortic Diameter manual measurement. WSS and circumferential component showed statistically significant association with local growth rate. Circumferential WSS showed statistically significant positive associations in area of aorta with fastest growth. [21] 32 BAV patients 28 aged-match healthy controls Both groups had baseline 4D-Flow CMR, BAV patients further had baseline/follow-up CTA (3 years later follow-up). Circumferential WSS and WSS angle higher in BAV group. WSS angle only parameter demonstrated association with aortic growth. [23] 72 BAV patients 136 age/sex matched controls 4D-Flow CMR for WSS assessment. MRI baseline and follow-up at ≥ 5-years for aortic diameter measurement. Section of ascending aorta exposed to increase WSS at baseline was increased for subjects with higher growth rates of compared to those with lower growth rates. Areas of elevated WSS in ascending aorta associated with elevated rates of dilatation. [56] Tortuosity 120 BAV subjects 234 TAV control subjects CT image analysis in 2D plane to calculate tortuosity. aortic arch tortuosity significantly increased in BAV group. [27] 120 BAV subjects 120 TAV controls BAV patients with CT scan retrospectively selected, arch tortuosity calculated by triangular distances in 2D images. Arch tortuosity significantly increased in BAV patients. Multivariate analysis total aortic tortuosity was significantly increased in 17 subjects with Type A dissection. [57] Reverse Flow 3 BAV subjects CTA of thoracic aorta then 3D surface models of aortic root created from scan images. Retrograde flow quantified via flow reversal ratio index. Peak systolic retrograde flow present in ascending aorta for all subjects. Flow reversal dependent on BAV morphology. [58] 510 BAV subjects 120 TAV subjects with mid-ascending aorta diameter > 35 mm 25 controls 4D-Flow CMR for reverse/forward flow analysis, use of in-house programme for calculations. Reverse flow in ascending aorta in BAV subjects without AS or AR showed 94 % increase compared to healthy controls p < 0.001. In BAV cohort, elevated systolic and diastolic reverse flow in the ascending aorta significantly associated with increased mid-ascending aortic diameter. [20] 73 BAV subjects 23 healthy controls 4D-Flow CMR with flow analysis in 5 regions of the aorta with maximum aortic diameter calculations. BAV patients demonstrated increased helical flow regions with prominent reverse flow direction in the ascending aorta. Severely dilated patients showed more reverse flow in the aortic arch compared to non-dilated subjects (106 %, p < 0.01). [19] WSS, Wall Shear Stress; 4D-CMR, 4-dimensional cardiovascular magnetic resonance imaging; CTA, computed tomography angiography; BAV, Bicuspid Aortic Valve; MRI, Magnetic resonance imaging; TAV, tricuspid aortic valve; AS, aortic stenosis; AR, aortic regurgitation. Table 2 Previously studied circulating biomarkers utilising BAV subjects. Biomarker Study Sample Method Key Results Reference s MMPs 8 studies with data on protein expression in thoracic aortic aneurysms involving BAV subjects also Meta-analysis, studies with thoracic aortic aneurysms with MMP/TIMP protein expression measurements in both these subjects and controls, also included studies where subjects with thoracic aortic aneurysms also had BAV. Increase in MMP-9 and no change in MMP-2 in subjects with thoracic aneurysms (n = 106) vs controls (n = 30) Highly significant increase in MMP-2 but not MMP-9 in thoracic aortic aneurysms in patients with BAV (n = 112) vs TAV (n = 53). BAV also showed TIMP-1 reduction compared to TAV group. [59] 16 BAV patients with thoracic aortic aneurysm 12 TAV patients with thoracic aortic aneurysm Aortic root and ascending aortic aneurysm tissue analysed, total activity of MMP-2, MMP-9 and TIMP-1/TIMP-2 levels were measured. Total MMP-2 activity higher in the walls of aneurysms in the BAV group vs TAV. No significant difference found in activity of MMP-9 in both groups. No significant differences in TIMP-1 + TIMP-2 between both groups. [54] 28 BAV patients 54 TAV patients Biopsy of aortic wall at time of surgery analysed, RT-PCT for gene expression of MMP-1, MMP-2, MMP-9, TIMP-1/TIMP-2. Greater ratio of MMP-2/TIMP-1 in BAV subset compared to TAV. No difference in gene expression levels of MMPs and TIMPs. [39] TGFβ1 317 patients with suspected or known aortic genetic syndrome. 30 BAV subjects with thoracic aortic aneurysm or dissection Total serum TGFβ1 measured in 317 subjects with known or suspected genetic aortic syndrome. Total serum TGFβ1 elevated in BAV vs patient without a genetic aortic syndrome. Across different genetic aortic syndromes BAV subjects had the highest levels of Total serum TGFβ1. [31] 50 patients total. 16 patients dilated aorta and BAV 14 patients dilated aorta and TAV 15 patients non-dilated aorta and BAV 5 patients non-dilated aorta and TAV 40 subject control group Aortic wall samples gained form 30 patients, Serum TGF-β1 levels quantified from blood samples collected. 2-fold decrease in serum TGF-β1 levels in patients with aortic dilation compared to reference group. In the BAV group, there was no significant difference in serum TGF-β1 levels between dilated and non-dilated aortas. No significant correlation between serum TGF-β1 levels and ascending aorta diameter. [32] miRNA 63 BAV patients with root dilatation 50 TAV patients without aortopathy as control Proximal aortic diameters assessed by transthoracic echocardiography, if diameter was ≥ 40.0 mm, patients underwent CT or MRI of thoracic aorta. 20 BAV aortopathy associated genes analysed. miRNA analysis via blood samples, selected miRNA were based on previous literature. Inverse linear correlation between miRNA-145 and ascending aorta diameter Strongest correlation between miRNA-17 + miRNA-106a Significant association between miRNA-145, miRNA-17, miRNA-106a and progression of aortic disease observed. [60] 63 BAV patients with AR 32 BAV patients with AS 50 TAV patients with severe TAV stenosis miRNA analysis of 11 miRNAs in 145 patients referred for AV surgery. TTE for assessment of proximal aortic diameter. miRNA-17 + miRNA-20a significant inverse linear correlation with maximum aortic diameter BAV AS subgroup, inverse correlation between maximum aortic diameter and miRNA-17 + miRNA-20a. [34] 20 BAV patients with aortopathy 4D-Flow CMR to assess areas of elevated and reduced WSS, aortic tissue removed from these areas during surgery. miRNA-34a, miRNA-125a and MMP-2 expression levels were quantified in tissues removed. miR-34a + miRNA125a down-regulated in elevated WSS areas. Possible inhibitors of MMP-2 gene via Luciferase reporter assay. miR-34a + miRNA125a may have involvement in WSS related aortopathy with their potential regulation of MMP-2. [61] Troponin and BNP 184 patients with BAV from 2 previous prospective observational cohort studies Venous sampling of Troponin T, NT-proBNP, CRP, TGF-β. TTE performed. Associations carried out between biomarker and arrhythmia-free/intervention free survival. Median follow-up of 6.9 years. Troponin-T significantly elevated in women compared to males. NT-proBNP levels significantly lower in men compared to females. Increased NT-proBNP showed strongest association with end point of arrythmias or death. [43] 183 patients with BAV from data of 2 previous observational studies. 145 healthy volunteers Subjects underwent TTE and venous sampling of several biomarkers (hsTnT + NT-proBNP) to assess associations with degree of AV stenosis/regurgitation and aortic diameter 145 healthy volunteers had normal TTE, age under 50 years old and had TGF-ß1 levels measured as aged matched controls. Increased hsTNT associated with increased risk of one higher grade of aortic regurgitation No biomarkers showed association with aortic diameter. [44] Desmosine 20 patients with BAV and healthy controls uDES and pDES measured via liquid chromatography-tandem mass spectrometry. Aortic root size and Z-score measured via TTE. 15 patients had BAV pDES and uDES significantly elevated in patients with BAV compared to controls Significant association between uDES and aortic root size and Z-Scores in the BAV subjects. [47] 239 patients with AAA recruited to the Ma3RS study ( NCT01749280 ) 6 month visits with abdominal ultrasound performed. Panel of biomarkers taken including pDES, patients followed up for events such as AAA rupture, repair, mortality. pDES significantly correlated with AAA diameter pDES major predictor of AAA events and MACE independent of AAA diameter. pDES significantly associated more with AAA events than MMP-9. [62] BAV, Bicuspid Aortic Valve; MMP, matrix metalloproteinase; TIMP, tissue inhibitors of metalloproteinase; TAV, tricuspid aortic valve; RT-PCT, Reverse transcription polymerase chain reaction; TGFβ1, Transforming growth factor beta 1; miRNA, MicroRNAs; CT, Computed Tomography; MRI, Magnetic resonance imaging; AS, aortic stenosis; AR, aortic regurgitation; AV aortic valve; TTE, transthoracic echocardiogram; WSS, wall shear stress; NT-proBNP, N-terminal pro b-type natriuretic peptide; hsTnT, high-sensitive troponin T; AAA, abdominal aortic aneurysm; MACE, major adverse cardiovascular event; pDES, plasma desmosine; uDES, urine desmosine. Previously studied imaging biomarkers utilising BAV subjects. WSS, Wall Shear Stress; 4D-CMR, 4-dimensional cardiovascular magnetic resonance imaging; CTA, computed tomography angiography; BAV, Bicuspid Aortic Valve; MRI, Magnetic resonance imaging; TAV, tricuspid aortic valve; AS, aortic stenosis; AR, aortic regurgitation. Previously studied circulating biomarkers utilising BAV subjects. BAV, Bicuspid Aortic Valve; MMP, matrix metalloproteinase; TIMP, tissue inhibitors of metalloproteinase; TAV, tricuspid aortic valve; RT-PCT, Reverse transcription polymerase chain reaction; TGFβ1, Transforming growth factor beta 1; miRNA, MicroRNAs; CT, Computed Tomography; MRI, Magnetic resonance imaging; AS, aortic stenosis; AR, aortic regurgitation; AV aortic valve; TTE, transthoracic echocardiogram; WSS, wall shear stress; NT-proBNP, N-terminal pro b-type natriuretic peptide; hsTnT, high-sensitive troponin T; AAA, abdominal aortic aneurysm; MACE, major adverse cardiovascular event; pDES, plasma desmosine; uDES, urine desmosine.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.