Bicuspid aortic valve (BAV) is one of the most common congenital cardiac abnormalities, occurring in 1-2% of the population [1]. A high heritability index [2] and association with variants in several genes that regulate heart development [3, 4] suggest that BAV is regulated by genetic factors. BAV is also associated with abnormalities of the thoracic aorta including dilation, aneurysm formation, coarctation, and dissection [5, 6]. More than half of young patients with normally functioning BAV have echocardiographic evidence of aortic dilatation [7], and progressive aortic dilation may develop in children [8]. The degree of aortic dilation appears out of proportion to valve dysfunction [9, 10], and aortic valve replacement does not halt the risk of subsequent aortic complications [7, 11]. Finally, aortic aneurysm can present in BAV families as an autosomal dominant trait with incomplete penetrance [12], and data suggests that genetic mutations may contribute to aortic aneurysm in the setting of a BAV [13]. Family studies, however, have not supported shared underlying genetic effects contributing to both BAV and aortic dilation [14]. Thus, evidence would support there being intrinsic abnormalities of aortic stiffness, perhaps caused by genetic factors that may invariably contribute to the risk of dilation and potential for aneurysm formation in BAV.

In this study we prospectively tested our hypothesis by measuring aortic stiffness in a cohort of BAV and TAV patients with normal proximal ascending aorta dimension. In order to further investigate vascular function in BAV, we also measured augmentation index (AIx) as a proxy of pressure from wave reflections and carotid-brachial/ carotid-radial pulse-wave velocity (cb-PWV) as indications of peripheral arterial stiffness.

This study demonstrates that in the setting of a normal ascending aorta size, aortic PWV and arterial elastance appear similar between subjects with BAV and those without. By comparison, BAV participants were found to have a significantly elevated AIx compared to controls. The increased pressure from wave reflections was independent of sex, height and heart rate [18–20]. This finding is clinically significant as increased AIx is associated with increased risk of future CV events [21].

We noted elevated AIx in BAV in the absence of elevated aortic PWV and this is novel. Previous studies noting increased AIx in BAV were carried out in patients with aortic dilation [22, 23]. This is important as patients with BAV and dilated aortas have increased aortic PWV compared to BAV patients with preserved aortic size [24]. Dilation of the vessel transposes load bearing from elastin fibers to stiffer collagen fibers, increasing the incremental elastic modulus of the vessel [25, 26]. According to the Moens-Korteweg equation, this increase in elastic modulus will increase PWV. Whether increased AIx was secondary to increased aortic stiffness (from altered wave speed) or an independent phenotypic expression of this unique pathology could not be disentangled. Our findings offer important insight into the arteriopathy associated with BAV and suggest increased pressure from wave reflections as a primary systemic vascular aberration in BAV in the absence of aortic dilation, subsequent stiffening and altered reflection timing. These findings suggest a role for downstream microvascular dysfunction distal to the aorta as a potential arbitrator of increased magnitude of pressure from wave reflections in BAV.

A novel observation in the present study was the association between AIx and peak wall shear offering insight into observations of higher AIx in the setting of BAV. BAV patients with normal aortic geometry had higher wall shear than TAV due to higher aortic flow velocity and this is consistent with previous reports noted in BAV patients with dilated aortas and/or aneurysms [27, 28]. Altered valve hemodynamics have been implicated in the cause of aortic root dilation [29, 30], aneurysm [31] and valve calcification [32, 33] in BAV. Such altered central hemodynamics [34] may also damage the peripheral microvasculature [35], altering reflection sites and contributing to increased reflected pressure wave magnitude. It is interesting to note that valve replacement and normalization of aberrant central flow profiles in BAV does not mitigate aortic dilation progression [11]. Abnormalities in AIx reported herein, and elsewhere [22, 23], may reflect a previously unsuspected role for increased pressure from wave reflections in the development of large vessel arteriopathy in patients with BAV. In other patient populations such as Marfan syndrome, increased AIx has been shown to predict progression of aortic disease [36, 37]. It has been suggested that pressure pulsatility and cyclic stress from wave reflections alter load-bearing capacity of the aortic wall and contribute to fatigue-fracture (i.e. mechanical failure of biomaterials), increasing risk for aortic dilation and aneurysm [38]. The etiology of AIx as it relates to aortopathy in BAV requires further investigation.

The results of our prospective analysis finding no differences in PWV and arterial elastance between BAV and TAV patients is supported by studies showing no difference in serum matrix metalloprotein-2 levels in BAV without aneurysms compared to normal controls [24]. In addition, gene expression profiles of aorta tissue taken from BAV and TAV patients without aneurysm are very similar [39]. Our results differ from studies that demonstrated abnormal aortic root distensibility and stiffness index in BAV patients without aneurysms [40–42]. Discrepancy is likely related to method of measurement as these previous studies relied on imaging modalities such as echocardiography. Our findings are in agreement with others [22, 24] demonstrating normal aortic stiffness as measured using PWV in BAV patients without aneurysms. PWV is considered to be a robust measure of aortic stiffness and is currently viewed as the “gold standard” for measuring aortic stiffness [18–20].

We acknowledge several limitations in our study. First, since our primary hypothesis was that abnormal aortic stiffness would be found in the setting of BAV we consider our observation of increased AIx to be hypothesis generating. Second, the small size of our study may limit the generalizability of our results to the broader population of patients with BAV. We noted a partial η² of 0.47 with an observed power of 0.79 signifying moderate effect size with adequate power for detecting group differences in AIx. Thus although possibility of a type II error exists, we believe it to be low.

Normal aortic stiffness with normal aortic dimension suggests that the ascending aorta is not invariably abnormal in the setting of BAV, despite the predisposition for aorta dilation and aneurysm formation. Our data should motivate further investigation into the impact of the peripheral vasculature on the central vascular phenotype in BAV patients as long term abnormalities may contribute to subsequent ascending aortic dilation and aneurysm formation in patients with BAV.

Our study demonstrates that arterial stiffness and elastance are similar between subjects with BAV and TAV with normal aortic dimensions. At the same time we demonstrate a higher carotid AIx in BAV patients consistent with abnormal vascular properties distal to the aorta. Future studies are required to determine whether abnormal distal arterial properties in BAV subjects contribute to proximal aortic dilation and aneurysm formation.