Summary
Background
Congenital bicuspid aortic valve (BAV) is the most common congenital heart defect and may be responsible for aortic stenosis early in life. However, its pathogenesis remains unclear. A relationship between the severity of aortic stenosis and valvular surfaces has not been reported in the paediatric population.
Aims
To assess the feasibility of three-dimensional transthoracic echocardiographic planimetry in congenital BAV in children and to evaluate the influence of valvular asymmetry and aortic valve area (AVA) on stenosis severity.
Methods
Seventy consecutive children with BAV were included in this prospective single-centre study. Using the multiplanar review mode, surfaces were measured by planimetry (in systole for AVA and diastole for cusp surfaces). The degree of stenosis was assessed by instantaneous aortic Doppler. Results are expressed as medians and first and third quartiles.
Results
Median age was 5.6 years (2.2–11.5). Feasibility was 97%. Intra- and interobserver concordances were excellent for the measurement of cusp surfaces and AVA. Among the 70 children, 25 had aortic stenosis. The small/large cusp ratio was strongly associated with aortic stenosis ( P < 0.001). The area under the receiver operating characteristic curve was 0.89 (95% confidence interval 0.82–0.97). The best cut-off value for differentiating stenotic from non-stenotic valve was 0.75, with 84% sensitivity and 83% specificity. When indexed for body surface area, AVA was significantly smaller ( P = 0.031) in case of stenotic BAV (1.51 cm 2 [0.99–2.28]) compared with non-stenotic BAV (1.99 cm 2 [1.57–2.52]).
Conclusions
Three-dimensional echocardiographic planimetry is a feasible and reproducible method for assessing aortic surfaces in congenital BAV. Aortic stenosis seems to strongly depend on the asymmetry of the valve.
Résumé
Contexte
La bicuspidie aortique (BAo) est la malformation cardiaque congénitale la plus fréquente. Elle peut être responsable de sténose aortique (SAo) dès le plus jeune âge. Cependant, sa pathogénie exacte demeure incertaine. La relation entre sévérité de la SAo et importance des surfaces valvulaires n’a jamais été montrée chez l’enfant.
Objectifs
Étude de la faisabilité de l’échocardiographie transthoracique tridimensionnelle (E3D) pour l’étude planimétrique des BAo de l’enfant et étude de l’influence de l’asymétrie valvulaire ainsi que celle de l’orifice valvulaire aortique sur la sévérité de la SAo.
Méthodes
Soixante-dix enfants ayant une BAo ont été inclus consécutivement dans cette étude prospective monocentrique. Grâce au mode de revue multi-plan, les surfaces valvulaires ont été mesurées par planimétrie (en systole pour l’orifice aortique et en diastole pour les feuillets). Le degré de SAo était estimé par le mode Doppler. Les résultats sont présentés sous forme de médiane, premier et troisième quartiles.
Résultats
L’âge médian était de 5,6 ans (2,2–11,5). La faisabilité était de 97 % avec une excellente variabilité intra- et interobservateur tant pour la mesure des feuillets valvulaires que pour celle de l’orifice aortique. 25 enfants avaient une SAo. Le ratio petit/grand feuillet était corrélé à l’existence d’une SAo ( p < 0,001), l’aire sous la courbe ROC étant de 0,89 (IC à 95 % : 0,82–0,97). La meilleure valeur seuil permettant de différencier une valve sténosante d’une non sténosante était de 0,75 (sensibilité 84 % et spécificité 83 %). Indexé à la surface corporelle, l’orifice valvulaire était significativement ( p = 0,031) plus petit en cas de BAo sténosante (1,51 cm 2 [0,99–2,28 vs 1,99 cm 2 [1,57–2,52]] pour les non sténosantes).
Conclusions
La mesure planimétrique des surfaces valvulaires par E3D est une technique faisable et reproductible en cas de BAo chez l’enfant. La SAo semble dépendre étroitement du caractère asymétrique de ces valves.
Background
Bicuspid aortic valve (BAV) is the most common congenital heart disease, observed in 0.4 to 2.25% of the general population . The clinical spectrum is wide, ranging from asymptomatic valvular disease to severe heart failure. The severity of the valve dysfunction is related to the morphology of the BAV . Moreover, the phenotype of the BAV is predictive of surgical timing . Thus, the challenge is to assess valvular function and determine its probable evolution. Three-dimensional echocardiography (3DE) was reported as an efficient means of assessing valvular morphology in BAV . Indeed, using multiplanar analysis, this technique allows precise measurement of valvular areas . In adults, real-time three-dimensional transthoracic echocardiography (RT-3DTTE) using the biplane mode was consistent with both transoesophageal 3DE and catheterization for measurement of the aortic valve area (AVA) . Although previous studies have demonstrated that the severity of aortic stenosis closely depends on the structural geometry of the aortic valve, the exact pathogenesis remains unclear. The aims of this study were to evaluate the feasibility of RT-3DTTE for the assessment of aortic surfaces in congenital BAV and to assess the relationship between valvular asymmetry, AVA and aortic stenosis.
Methods
Definition of bicuspid aortic valve
The term BAV includes different morphological phenotypes. Indeed, the valve may be composed of either two cusps (purely bicuspid) or three cusps (falsely bicuspid) with fusion (partial or complete) of two of them. The term raphe defines the conjoint or ‘fused” area of the two underdeveloped cusps. For this study, we used the anatomical classification proposed by Sievers and Schmidtke , which is based on the presence and position of the raphes. Thus, a BAV type 0 (true BAV) has only two cusps with no raphe and BAV types 1 and 2 have three cusps with one and two raphes, respectively.
Patients
This was a single-centre prospective study performed in our paediatric cardiology unit. From April 2010 to April 2011, we enrolled all children with an echocardiographic diagnosis of BAV. These children were initially referred for the exploration of a cardiac murmur or for follow-up of an already known congenital heart disease. Exclusion criteria were age > 18 years, history of endocarditis, concomitant subaortic obstruction, previous surgical aortic commissurotomy and previous percutaneous aortic dilation. Only BAV types 0 and 1 (with 0 raphes and one raphe, respectively) were included because BAV type 2 (three cusps and two raphes) may be considered as a functional unicuspid aortic valve. The study was approved by our local ethics committee. A written consent form was not required according to French law, given that echocardiographic evaluation was part of the regular management of the children and was required for management of their medical conditions. No additional examination was performed for the sole purpose of the study.
Echocardiographic acquisitions and off-line analysis
All patients underwent standardized and complete two-dimensional transthoracic echocardiography (2DTTE), performed using high-quality commercially available ultrasound systems (iE33, Philips Medical Systems, Andover, MA, USA). When a BAV was diagnosed, the examination was completed by RT-3DTTE. X3-1 and X7-2 matrix probes (Philips Medical Systems, Andover, MA, USA) were used, depending on the age of the patient. The best short-axis cross-sectional view was selected to see the most circular annulus and the commissural areas. RT-3DTTE data were acquired–as much as possible considering the extreme youth of some children – during end expiratory. When necessary, 3D zoom was used to improve image resolution. Q-Lab software (Philips Medical Systems, Andover, MA, USA) was used for off-line analysis. In the multiplanar review (MPR) mode, three orthogonal cutting planes can be moved independently of each other through the data set. The MPR mode allows better morphological analysis because the valve can be seen in the best 2D cutting plans at any stage of the cardiac cycle. 3D images were acquired by triggering to the electrocardiogram R-wave and quality was improved using gain and compression controls. Two 3D volumes were acquired for each patient. Cusp surfaces and AVA were determined by planimetry. The degree of asymmetry was assessed by manually surfacing the leaflets ( Fig. 1 ). Aortic cusp surfaces were measured in diastole. For BAV type 1, the sum of the two fused cusps was considered as a single functional cusp. The ratio between the surfaces of the two functional cusps was calculated to express the degree of asymmetry (small/large cusp ratio). AVA was measured at the inner leaflet edges at the time of maximal opening in systole. All areas were corrected for body surface area. Mean and maximal instantaneous aortic Doppler gradients were measured to evaluate aortic stenosis. Aortic Doppler values were obtained from an apical five-chamber view. The valve was defined as stenotic for a maximal instantaneous aortic Doppler gradient > 20 mmHg. The ascending aorta, measured in parasternal long-axis view, was defined as dilated if the Z-score of the diameter was > 2.
Operators
RT-3DTTE acquisitions were performed by a single cardiologist. Two independent operators performed the off-line analysis, randomly and on separate days, to assess interobserver variability. To evaluate intraobserver variability, the same observer performed measurements twice, on different days, in a random order and blinded to the prior results.
Statistical analysis
Statistical analysis was performed on STATA statistical software, version 11.1 (Stata Corporation, College Station, TX, USA). Continuous variables are summarized as medians and interquartile ranges. Categorical variables are presented as proportions. In univariate analysis, qualitative variables were compared with the Chi 2 test (or Fisher’s exact test; when necessary). The Mann–Whitney U test was used to compare ranges of continuous non-normally distributed variables according to qualitative variables. Reproducibility of the measurements was tested on 16 randomly selected patients by calculating intraclass correlation coefficients. Receiver operating characteristic curves were used to determine the best threshold values for sensitivity and specificity. Linear regression analyses expressing Doppler gradients according to AVA or small/large cusp ratio are presented. A P value < 0.05 was considered statistically significant.
Results
Population characteristics
Among the 70 patients included in the study, 44 (62.9%) were male. Median age was 5.6 years (2.2–11.5) and median weight was 16.5 kg (11–32) ( Table 1 ). The most common associated cardiac lesion was coarctation of the aorta, encountered in 21 (30%) patients ( Table 2 ). Aortic stenosis was noted in 25 patients (35.7%). The median value of the mean instantaneous aortic Doppler gradient was 16 mmHg (10–30) for stenotic BAV and 2 mmHg (2–3) for non-stenotic valves. The median value of the maximal instantaneous aortic Doppler gradient was 30 mmHg (24–50) in case of aortic stenosis and 4 mmHg (4–5) in case of non-stenotic BAV. Dilation of the ascending aorta occurred in 22 patients (31.4%) and was more frequent in case of aortic stenosis (48.0% vs. 22.2%; P = 0.03). Four patients were excluded from the study because they had BAV type 2; three of them had aortic stenosis (one patient underwent a surgical commissurotomy). According to standard cross-sectional echocardiography, all patients had normal left ventricular function.
