To the Editor:

Is the aortic dilation that is commonly seen with bicuspid aortic valve (BAV) related to intrinsic aortic wall fragility or altered systolic hemodynamics? Recent publications on the topic favor the intrinsic fragility hypothesis. But recent advancements in imaging show very abnormal systolic flow in the ascending aortas of patients with BAV (1). We have been following a cohort of such patients who have eccentric, helical flow as demonstrated by a cardiac magnetic resonance (CMR) technique for dynamic blood flow imaging termed 4D Flow. One consequence of this eccentric flow is focally elevated wall shear stress at the convexity of the ascending aorta (2). We hypothesize that the degree of eccentric flow and resultant skewed shear stress directly contribute to the development of aortic pathology. Others have examined eccentric flow and enlarged aortic dimensions at a single time point (3), but causality and temporal evaluation have not been addressed. Here we investigate the relationship between flow and ascending aortic growth rates in patients with BAV.

Serial magnetic resonance or computed tomography angiography studies were retrospectively reviewed in 25 patients evaluated with 4D Flow. All studies were performed for clinical indications. Thirteen patients had BAV (26.5 years of age, 5 female, 3.5 ± 0.7 cm baseline aortic size), and 12 with tricuspid aortic valve (TAV) served as controls (30.7 years of age, 3 female, 3.4 ± 0.5 cm baseline aortic size). The patients with BAV were relatively free of significant aortic valve disease; only 3 had greater than mild stenosis and 1 greater than mild insufficiency. None of the patients with TAV had valve disease. Ascending aortic diameters were measured at standard levels by 2 blinded reviewers, and the growth rate of the maximally enlarged segment was determined. Using a previously described qualitative approach, ascending aortic blood flow was graded as normal or mildly or markedly eccentric (2). All TAV controls had normal flow, whereas 10 of the 13 patients with BAV had abnormal flow (7 markedly eccentric). Normalized flow displacement from the vessel center, a recently developed parameter to quantify flow eccentricity, was calculated as well (4). Growth rates for controls and patients with BAV, and within the BAV group based on the degree of eccentric flow, were compared with unpaired Student t tests. Growth rates were also correlated with the flow displacement variable using a linear regression model.

The average follow-up was 4.3 ± 2.9 years. The growth rates of patients with BAV were significantly higher than those of TAV controls (0.8 ± 0.7 mm/year vs. 0.1 ± 0.2 mm/year, p value = 0.004). Among patients with BAV, those with abnormal flow patterns demonstrated significantly higher growth rates than those with normal flow (1.0 ± 0.6 mm/year vs. 0.0 ± 0 mm/year, p value = 0.02). Furthermore, patients with BAV demonstrating markedly eccentric flow exhibited more rapid growth than other patients with BAV (1.2 ± 0.7 mm/year vs. 0.3 ± 0.4 mm/year, p value = 0.02). Linear regression between normalized flow displacement and aortic growth rate showed good correlation (r = 0.66).

These results suggest that altered systolic hemodynamics in the ascending aorta related to BAV play a more significant role in aortic dilation than currently believed. Higher growth rates are seen in patients with BAV and abnormal flow (Figure 1). Patients with markedly eccentric flow demonstrate the most rapid growth. Two such patients in our cohort have since required surgical aortic repair. Good correlation is found between growth and a quantitative measure of flow eccentricity, normalized flow displacement. Our results complement recent data from Della Corte et al. (5) that demonstrate increased growth rates with restrictive leaflet excursion (a surrogate for flow eccentricity) in patients with BAV and right-left aortic leaflet fusion.

Imaging biomarkers such as these could be used to identify and risk-stratify patients in whom clinically significant aortic disease is likely to develop. We have shown how both qualitative and quantitative CMR markers of altered hemodynamics correlate with disease progression. These biomarkers are uniquely assessed by CMR. A sizable portion of patients with BAV and eccentric systolic flow do not have any measurable aortic stenosis (2), including 4 of 10 patients in this study, and thus would not be identified as at risk of post-stenotic dilation by echocardiography. Moreover, the 3-dimensional assessment of wall shear stress and flow displacement that we report is only achieved with multidirectional phase-contrast CMR imaging.

Yet does the correlation that we report imply causation? Are the alterations in systolic flow that we have linked to increased aortic growth merely the consequence of dilated aortic geometries? We think not because we have shown the same altered flow patterns in normal-sized aortas and have connected unique flow patterns to different aortic leaflet fusion patterns with BAV (1). The hemodynamic mechanism that we propose for progressive aortic growth will, of course, need to be validated with larger, prospective studies that carefully control for confounding variables. But the fact that hemodynamic biomarkers can be imaged noninvasively and potentially identify aortic disease well before clinical manifestation is an attractive goal that could meaningfully improve the clinical management of patients with aortic valve disease. It may also point to unique, pathologic cellular mechanisms that could be the targets of new medical therapy.