Subjects

We retrospectively screened patients with diagnoses of BAV in the echocardiographic database of the cardiac imaging laboratory at Asan Medical Center. Patients aged >19 years who had undergone aortic valve (AV) surgery between 2000 and 2014 were initially eligible ( n = 791; Figure 1 ). We excluded 52 patients who were confirmed to have tricuspid AVs during surgery by the cardiac surgeons. Additionally, because the chronic clinical consequences associated with BAV were our main interest, we excluded 58 patients who underwent surgery for the management of active infective endocarditis. Therefore, a total of 681 patients were included in the study. Using all available transthoracic echocardiographic (TTE), transesophageal echocardiographic, and computed tomographic images, BAV phenotypes were confirmed by consensus of three expert cardiologists (B.J.S., S.L., and J.-K.S.), each of whom has >10 years of clinical experience. The type of dominant valvular dysfunction (aortic stenosis [AS] vs aortic regurgitation [AR]) requiring AV surgery was also determined by consensus of the three experts using echocardiographic data and the recently published guidelines for the management of valvular heart disease. The presence of significant aortopathy associated with BAV was defined as a requirement for aortic surgery, including aortic replacement, reconstruction, and wrapping at the time of AV surgery. The study conformed to the ethical guidelines of the Declaration of Helsinki. The study protocol was approved by the ethics committee of Asan Medical Center, and the informed consent was not required.

Flowchart showing enrollment of subjects ( left ) and imaging modalities used for phenotype classification ( right ). CT , Computed tomography; TEE , transesophageal echocardiography.

BAV Phenotype Classification

Figure 2 shows a conventional and simple dichotomous method for the classification of BAV phenotypes. The conventional classification depends on the identification of each fused cusp with a demonstration of raphe. Thus, type 1 represented the fusion of the right coronary cusp (RCC) and the left coronary cusp (LCC), with or without raphe. Types 2 and 3 represented the fusion of the RCC/noncoronary cusp (NCC) and the LCC/NCC, respectively, with definite raphe. Type 4 included patients with BAV whose imaging data suggested the fusion of the NCC with the RCC or the LCC but failed to demonstrate raphe to adequately classify the subtype. A dichotomous phenotypic classification was developed by Jilaihawi et al. , which was based on the orientation of fused cusps and their relationship with coronary ostium. Thus, the coronary cusp fusion (CCF) type represented the fusion of the RCC and the LCC, showing the anterior-posterior orientation of the BAV with both coronary arteries arising from the fused anterior cusp, while in the mixed cusp fusion (MCF) type, the fusion of the RCC/NCC or the LCC/NCC resulted in right-left orientation of the BAV, with each coronary artery arising from each RCC and LCC. Figure 3 shows representative images according to the BAV phenotype and dominant valvulopathy.

Diagram of two different BAV phenotype classification methods. The conventional classification method is based on the identification of the individual cusps that are fused and the position of raphe ( left ), and the new dichotomous classification method is based on the orientation of fused cusp and coronary ostium ( right ).

Representative images of BAV with different phenotypes and patterns of valvulopathy. CT , Computed tomography; L , LCC; LCA , left coronary artery; N , NCC; R , RCC; TEE , transesophageal echocardiography.

Performance of the BAV Classification Method

To compare the clinical performance of different BAV classification methods (conventional vs simple dichotomous classification), we randomly selected TTE images of 100 patients. For application of simple dichotomous classification, the orientation of the fused cusp of the parasternal short-axis echocardiographic image was the primary information used. In some cases showing atypical or deviated orientation, we used the spatial relationship of BAV in relation to the adjacent cardiac structures, including the right ventricular outflow tract as well as the position of coronary ostium (representative images in Figure 4 ). Four cardiologists with different levels of clinical experience participated; cardiologists A and B had been dedicated to echocardiography for >5 years each, whereas cardiologists C and D had <2 years’ experience. They individually finalized the phenotypic classification using both the conventional and the dichotomous methods. TTE imaging was the only modality used for the comparison of performance, as it is accepted as a standard imaging technique for the screening and diagnosis of BAV. If the images were suboptimal and adequate classification was not possible, the final decision was reported to be unclassified. The frequency of infeasibility was compared among the reviewers. Using all cases with feasible classification, the accuracy of each classification method was determined on the basis of consensus phenotypes.

Representative images of applying a dichotomous classification. Parasternal short-axis TTE images showing how to apply a dichotomous classification for BAV. The orientation of the fused cusp is the primary information we used in patients with typical orientation (A,B) . In some patients with atypical or deviated orientation (C,D) , the spatial relationships to the adjacent structures and coronary ostium are helpful. In this case, the left coronary ostium ( arrow ) can locate the LCC (L) easily. The adjacent cusp facing the right ventricular outflow tract (RVOT) should be the RCC (R), and thus we can reliably diagnose fusion of the RCC and LCC resulting in BAV CCF type. N , NCC.

Data Analysis

All categorical variables are expressed as numbers with percentages, and all continuous variables are expressed as mean ± SD. For the comparison of variables, we used the χ ² test, Fisher exact test, and Student’s independent t test, as appropriate. Using logistic regression, we assessed the associations between baseline variables, including BAV phenotypes, and the pattern of valvulopathy. Variables with P values < .10 in the univariate models were tested in the multivariate model. We also assessed the associations between baseline variables and concomitant aortic surgery using the same method. For the validation of the TTE classification, we assessed the agreement between raters using the κ coefficient, and we determined the feasibility and accuracy of the dichotomous and conventional classifications. All reported P values were two sided, with a P value of <.05 indicating statistical significance. SPSS version 22 (SPSS, Chicago, IL) was used for statistical analysis.

Distribution of BAV Phenotypes

The dichotomous classification identified BAV CCF in 361 patients (53%) and BAV MCF in 320 (47%). All patients with CCF were classified as type 1 in the conventional BAV classification, whereas the 320 patients with MCF were classified as type 2 ( n = 193), type 3 ( n = 12), and type 4 ( n = 115), respectively ( Figure 5 , top ).

Distribution of BAV phenotypes and their association with clinical consequences (dominant aortic stenosis and concomitant aortic surgery). OR , Odds ratio.

BAV CCF versus BAV MCF

Table 1 presents demographic and laboratory data for the subjects and compares these clinical features according to BAV phenotype. Among a total of 681 patients (mean age, 59 ± 12 years; 424 men), the prevalence rates of hypertension and diabetes were 37% and 12%, respectively. Echocardiography showed that moderate to severe AS was more frequently observed than moderate to severe AR, regardless of BAV type. However, the frequency of moderate to severe AS was greater in the patient group with the BAV MCF type than in those with the BAV CCF type (91% vs 75%). Patients with the BAV CCF type, showing a relatively higher frequency of moderate to severe AR, showed a larger left ventricular (LV) cavity and LV mass index with a slightly lower LV ejection fraction. Patients with the BAV CCF type showed a larger diameter in the LV outflow tract and sinus of Valsalva, whereas those with BAV MCF showed a larger diameter in the tubular portion of the ascending aorta. Although only the BAV CCF type was identified in three patients with coarctation of the aorta, the frequency was not statistically different between the types.

The surgical procedures are summarized in Table 2 . Of a total of 681 patients, 546 (80%) underwent AV surgery for AS-dominant valvulopathy. In the BAV MCF group, the frequency of AS-dominant valvulopathy was 89%, which was higher than in the BAV CCF group (73%). Concomitant aortic surgery was performed in 214 patients (31%); this happened more frequently in the BAV MCF group than in the BAV CCF group (38% vs 26%, P = .002). The analysis of the causes of aortic surgery revealed that aortic aneurysm made up most cases ( n = 208 [97%]), with aortic dissection ( n = 5) and coarctation of the aorta ( n = 1) being rarely noted. Their distributions were not different between the groups.

Association Between BAV Phenotype and Valvulopathy and Aortopathy

Univariate analyses revealed that factors associated with AS-dominant valvulopathy included age, gender, body surface area, blood pressure, diabetes, and BAV MCF type ( Table 3 ). After adjusting covariates with the multivariate model, the BAV MCF type was found to be an independent variable associated with AS-dominant valvulopathy (odds ratio, 3.32; 95% CI, 1.99–5.54; P < .001; Figure 5 , middle ). For combined aortic surgery, diastolic blood pressure, diabetes, coronary artery disease, atrial fibrillation, and BAV MCF type were found to be significant. In the multivariate model, BAV MCF type was an independent factor associated with concomitant aortic surgery (odds ratio, 1.76; 95% CI, 1.26–2.45; P = .001; Figure 5 , lower ).

Table 3

Variables associated with AS-dominant valvulopathy and aortopathy

	Unadjusted			Adjusted
	OR	95% CI	P	OR	95% CI	P
Association with AS-dominant valvulopathy
Age	1.12	1.10–1.14	<.001	1.11	1.09–1.13	<.001
Male gender	0.20	0.12–0.33	<.001	0.31	0.17–0.55	<.001
Body surface area	0.03	0.01–0.11	<.001	1.97	0.37–10.54	.429
Systolic blood pressure	0.98	0.97–0.99	<.001	0.98	0.97–1.00	.020
Diastolic blood pressure	1.05	1.03–1.07	<.001
Diabetes	5.46	1.96–15.19	.001	4.43	1.42–13.80	.010
Coronary artery disease	1.55	0.85–2.83	.153
BAV MCF	2.98	1.97–4.52	<.001	3.32	1.99–5.54	<.001
Association with combined significant aortopathy
Age	1.01	0.99–1.02	.413
Male gender	1.05	0.75–1.47	.764
Systolic blood pressure	1.00	0.99–1.01	.315
Diastolic blood pressure	1.02	1.00–1.03	.020	1.02	1.00–1.03	.045
Diabetes	0.53	0.31–0.94	.028	0.56	0.32–1.00	.050
Coronary artery disease	0.49	0.29–0.83	.008	0.49	0.28–0.83	.009
Atrial fibrillation	1.82	1.09–3.05	.023	2.01	1.17–3.43	.011
BAV MCF	1.70	1.23–2.36	.001	1.76	1.26–2.45	.001
AR dominance	0.94	0.63–1.42	.768

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Looking Back, Looking Ahead, Revisited
2. Pediatric Appropriate Use Criteria for Outpatient Echocardiography: Practice Variations among Pediatric Cardiologists, Noncardiologist Subspecialists, and Primary Care Providers
3. Global and Regional Right Ventricular Function Assessed by Novel Three-Dimensional Speckle-Tracking Echocardiography
4. Application of Speckle-Tracking Echocardiography in an Experimental Model of Isolated Subendocardial Damage

Stay updated, free articles. Join our Telegram channel

Join