Background
Bicuspid aortic valves (BAVs) are associated with aortopathy. Recent studies suggest that aortic dilatation is more likely to be seen with left-right coronary cusp fusion (type I) compared with right-noncoronary cusp fusion (type II). The aim of this study was to investigate the association between BAV morphology and patterns of aortopathy.
Methods
Aortic dimensions and BAV morphology were obtained retrospectively from archived cine loops of 581 consecutive patients with BAVs and 277 matched normal controls from the Vancouver General Hospital echocardiography database. Patient demographics and other echocardiographic parameters were extracted from the database.
Results
The study population was composed of 71% type I BAVs (415 patients) and 26% type II BAVs (149 patients). Aortic dilatation was present in 30% of the population. Type I BAV was associated with increased dimensions indexed to body surface area at the sinus of Valsalva compared with type II BAV. No difference in proximal ascending aortic dimension was seen between different BAV morphologies. The pattern of dilatation with type I BAV was more likely to be at the level of the annulus or sinus of Valsalva compared with type II BAV (62% vs 33%, P = .002). Type I BAV was an independent predictor of proximal aortic dilatation (odds ratio, 3.42; 95% confidence interval, 1.07–10.9).
Conclusions
Type I BAV is associated with a greater likelihood of dilatation at the annulus and sinus of Valsalva. There is relative sparing of this region of the aorta in patients with type II BAVs. Individuals with different BAV morphologies may require different strategies of aortopathy surveillance.
Bicuspid aortic valves (BAVs) are the most common congenital abnormality, estimated by echocardiographic and autopsy studies to occur in 0.5% and 1% to 2% of the general population, respectively. Although they were once thought to be a benign condition, BAVs are now recognized to be associated with a number of complications, including valvular stenosis, regurgitation, infective endocarditis, and abnormalities of the thoracic aorta.
A number of abnormalities of the aorta can be seen with BAV, including aortic dissection, dilatation, aneurysm, coarctation, and an interrupted aortic arch. Although aortic dissection is always the most feared of these vascular complications, its precursor, aortic dilatation, is by far the most common and has been estimated to occur in approximately 40% to 60% of those with BAVs. Aortic dilatation may involve the aortic root (proximal to and including the sinotubular junction), the proximal ascending aorta above the sinotubular junction, or both.
Two bicuspid phenotypes are commonly described: type I (“anterior-posterior” valve opening), which results from fusion of the right and left coronary cusps, and type II (“right-left” valve opening), which results from fusion of the right and noncoronary cusps. A third phenotype, type III, results from the fusion of the left and noncoronary cusps but is exceedingly rare in comparison with the other two. Existing studies in the published literature have provided discrepant results, with some showing increased aortic dimensions with type I BAV, some with type II BAV, and yet others showing no association at all.
Given these conflicting results, we sought to more definitively characterize whether an association exists between the morphology of the BAV and the degree of aortic dilatation.
Methods
Study Population
Adult patients identified as having BAVs on transthoracic echocardiography from May 2003 to November 2009 were retrospectively identified from the Vancouver General Hospital echocardiography database. Patients were excluded if there was evidence of previous aortic valvuloplasty, corrective aortic root surgery, heart transplantation, complex congenital heart disease (except for coarctation of the aorta), or aortic valve endocarditis. When serial echocardiographic studies for an individual patient were available, only the most recent study was selected for quantitative aortic root measurements.
Selection of Control Group
A control group was obtained from the cohort of patients with normal results in the echocardiography database. The group was selected in a 1:2 ratio for every BAV patient and was matched for age, sex, and body surface area (BSA). Normal aortic dimensions were defined in the database according to published values in the literature, including an aortic annulus ≤32 mm, sinus of Valsalva ≤40 mm, sinotubular junction ≤35 mm, and proximal ascending aorta ≤38 mm.
Analysis of Echocardiographic Cine Loops
Assessment of BAV morphology and measurement of aortic root and proximal ascending aortic dimensions were performed on archived digital cine loops by a single investigator (C.K.), with a second investigator (J.J.) available to resolve any ambiguities. Aortic dimensions in the control group were assessed independently by a third investigator (C.C.). Measurements of the aorta were obtained at the level of the aortic annulus, sinus of Valsalva, and sinotubular junction as defined by the American Society of Echocardiography. The proximal ascending aorta, if visualized on available cine loops, was measured ≥1 to 2 cm distal to the sinotubular junction. Measurements were obtained in end-diastole and were taken from inner edge to inner edge, as recommended by published guidelines. All aortic measurements were obtained blinded to the BAV morphology. Cut points for pathologic aortic dilatation were defined as the mean + 2 SDs of aortic dimensions measured in the control group.
The presence of BAV was confirmed with the visualization of two fused aortic cusps in both systole and diastole in the parasternal short-axis view. BAV morphology was classified into the following three classes: fusion of the right and left coronary cusps (type I), fusion of the right and noncoronary cusps (type II), and fusion of the left and noncoronary cusps (type III). When BAV morphology assessment was limited by suboptimal image quality, images from previous serial studies were used to help resolve any uncertainties.
Statistical Analyses
All analyses were performed using SPSS version 19.0 (SPSS, Inc., Chicago, IL) and Excel 2007 (Microsoft Corporation, Redmond, WA). Continuous data are reported as mean ± SD. Categorical data are reported as percentages. Comparison of aortic dimensions was performed with one-way analysis of variance or independent two-sample Student’s t tests, as applicable. For comparison of dichotomous or categorical values, χ 2 or Fisher’s exact tests were used. Post hoc comparisons were performed using Tukey’s range test. Significant univariate predictors were included for multivariate analysis. Individuals with missing data were excluded when the measurement in question was required for the specific computation.
Reproducibility of Aortic Measurements
Intraobserver and interobserver variability was determined using a random selection of 25 representative studies from the study pool. Aortic measurements were obtained twice by the primary investigator (C.K.) on two separate days and repeated by a second experienced investigator (J.J.). The mean measurement at each site was determined, and the percentage deviation from the mean was used to estimate measurement variability. Intraobserver variability across all aortic sites was 3.6%, and interobserver variability was 6.1%.
Results
Baseline Characteristics
A total of 581 consecutive patients with BAVs were identified and analyzed from the echocardiography database. There were 71% type I (415 patients), 26% type II (149), and 3% type III (17) BAVs. Because of the small number of patients with type III BAVs, these individuals were excluded from further analyses, yielding a final study population of 564 patients.
The mean age was 49 ± 15 years, with 71% of patients with BAVs being men. Aortic coarctation was rare, with only 11 patients (2%) having echocardiographic evidence of the condition. There was no significant difference in age, gender, or the incidence of aortic coarctation between those with type I and type II BAVs ( Table 1 ). Patients with type I BAVs had greater BSAs compared with those with type II BAVs (1.91 ± 0.24 vs 1.85 ± 0.24 m 2 , P = .01).
| Variable | Type I BAV ∗ ( n = 415) | Type II BAV † ( n = 149) | Control ( n = 277) | P ‡ |
|---|---|---|---|---|
| Age (y) | 49.1 ± 15.1 | 46.8 ± 14.3 | 48.3 ± 8.4 | .18 |
| Men | 302 (73%) | 97 (65%) | 200 (72%) | .19 |
| BSA (m 2 ) | 1.91 ± 0.24 | 1.85 ± 0.24 §|| | 1.90 ± 0.13 | .01 |
∗ Fusion of right and left coronary cusps.
† Fusion of right and noncoronary cusps.
‡ Results of between-group differences with one-way analysis of variance.
The control group consisted of 277 patients matched for age, sex, and BSA with normal results on echocardiography. Although there were no statistical differences in age and sex compared with the study group, BSAs in the control group remained statistically larger than in patients with type II BAVs. There was no difference in BSA between the control group and either patients with type I BAVs or the entire BAV cohort.
Echocardiographic Characteristics
The majority of patients in our study were free of hemodynamically significant valvular disease, with only 22% (68 of 312 patients) with severe aortic stenosis (AS) and 4% (20 of 563 patients) with severe aortic regurgitation. Baseline echocardiographic characteristics between the two BAV morphologic subtypes are displayed in Table 2 . There were no significant differences in measures of AS severity between the two groups as assessed by transvalvular pressure gradients, aortic valve area, or the proportion of individuals with severe AS (aortic valve area <1.0 cm 2 ).
| Variable | Type I BAV ∗ ( n = 415) | Type II BAV † ( n = 149) | P |
|---|---|---|---|
| Relative wall thickness | 0.40 ± 0.09 | 0.43 ± 0.30 | .17 |
| LV mass (g) | 183 ± 64 | 178 ± 68 | .35 |
| LVEF (%) | .06 | ||
| >50 | 385 (93%) | 136 (91%) | |
| 35–50 | 14 (3%) | 10 (7%) | |
| 20–35 | 13 (3%) | 1 (1%) | |
| <20 | 3 (1%) | 2 (1%) | |
| Peak gradient (mm Hg) | 30 ± 26 | 33 ± 18 | .35 |
| Mean gradient (mm Hg) | 17 ± 13 | 18 ± 10 | .31 |
| AVA (cm 2 ) | 1.4 ± 0.6 | 1.4 ± 0.5 | .22 |
| Indexed AVA (cm 2 /m 2 ) | 0.76 ± 0.28 | 0.75 ± 0.28 | .79 |
| Severe AS (AVA <1 cm 2 ) | 45 (21%) | 23 (23%) | .77 |
| More than moderate AR (%) | 15 (4%) | 5 (3%) | .58 |
∗ Fusion of right and left coronary cusps.
Aortic Dilatation and BAV Morphology
Compared with the control group, patients with type I BAVs had significantly larger dimensions at all levels when indexed to BSA ( Table 3 ). There appeared to be relative sparing of the proximal aorta in patients with type II BAVs, with no statistically significant difference in indexed dimensions of the sinus of Valsalva compared with the control group. These observations held true with nonindexed dimensions, although there was no significant difference in nonindexed annular dimensions in either BAV subgroup compared with controls.
| Dimension | Type I BAV | Type II BAV | Control | P ∗ |
|---|---|---|---|---|
| Nonindexed (mm) | ||||
| Annulus | 22.2 ± 3.3 (415) | 21.9 ± 3.3 (149) | 21.7 ± 2.1 (277) | .06 |
| Sinus of Valsalva | 34.1 ± 5.8 (415) † | 30.9 ± 5.0 (149) ‡ | 31.8 ± 3.7 (277) | <.001 |
| Sinotubular junction | 30.2 ± 5.6 (408) † | 28.9 ± 4.9 (149) †‡ | 27.6 ± 3.3 (277) | <.001 |
| Ascending aorta | 34.1 ± 7.0 (319) † | 32.9 ± 5.9 (110) † | 29.0 ± 3.6 (258) | <.001 |
| Indexed (mm/m 2 ) | ||||
| Annulus | 11.7 ± 1.6 (408) † | 11.9 ± 1.8 (146) † | 11.4 ± 1.1 (277) | .001 |
| Sinus of Valsalva | 18.0 ± 2.8 (408) † | 16.9 ± 3.0 (146) ‡ | 16.8 ± 1.9 (277) | <.001 |
| Sinotubular junction | 15.9 ± 2.9 (402) † | 15.8 ± 2.8 (146) † | 14.6 ± 1.7 (277) | <.001 |
| Ascending aorta | 17.9 ± 3.7 (316) † | 17.8 ± 3.5 (108) † | 15.3 ± 1.9 (158) | <.001 |
∗ Results of between-group differences with one-way analysis of variance.
Patients with type I BAVs had greater dimensions compared with those with type II BAVs ( Figure 1 ) at the level of the sinus of Valsalva (34.1 ± 5.8 vs 30.9 ± 5.0 mm, P < .001) that persisted after indexing to BSA (18.0 ± 2.8 vs 16.9 ± 3.0 mm/m 2 , P < .001) ( Table 3 ). These differences in indexed aortic dimensions are depicted graphically in Figure 2 . Individuals with any pathologic aortic dilatation (defined as greater than the mean dimension of the control group + 2 SDs) were older, had larger BSAs, and were more likely to be male than those without aortic dilatation ( Table 4 ). Nonindexed aortic dimensions of this subgroup of patients again consistently demonstrated greater sinus of Valsalva dimensions in type I versus type II patients ( Table 5 ).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
