Summary
Background
Accurate evaluation of aortic root geometry is necessary in congenital aortic valve lesions in children, to guide surgical or angiographical intervention.
Aim
To compare aortic annulus diameters measured by two- and three-dimensional transthoracic echocardiography (2D- and 3D-TTE), to determine the feasibility and reproducibility of 3D imaging and assess the dynamic changes during the cardiac cycle.
Methods
Thirty children without heart disease were prospectively included. Two orthogonal aortic annulus diameters were measured offline using multiplanar reconstruction in diastole and in systole and were compared with the measurement of the aortic annulus diameter by 2D-TTE.
Results
Mean age was 11 ± 3.6 years. Feasibility of 3D imaging was 100%. The coefficients of intra- and interobserver variability were 3.5% and 6%, respectively. The 2D mean diameter was significantly smaller than the 3D maximum diameter in systole (1.94 vs. 2.01 mm; p = 0.005). 2D and 3D measurements were well correlated ( p < 0.0001). The maximum and minimum diameters in 3D were significantly different both in systole and in diastole ( p < 0.001) underlining an aortic annulus eccentricity. The mean aortic annulus diameters were not significantly different between systole and diastole, with important individual variability during the cardiac cycle.
Conclusion
This study demonstrated the feasibility and reproducibility of 3D-TTE for the assessment of the aortic annulus diameter in a normal paediatric population. Because of an underestimation of the maximum diameter by 2D-TTE and the asymmetry of the aortic annulus, 3D measurements could be important before percutaneous aortic valvuloplasty or surgical replacement.
Résumé
Contexte
L’évaluation de la géométrie de la racine aortique est indispensable avant toute chirurgie ou cathétérisme interventionnel de la valve aortique.
Objectif
Le but de l’étude a été de comparer le diamètre aortique par échocardiographie transthoracique, de déterminer la faisabilité and la reproductibilité de l’imagerie 3D et d’analyser la variation au cours du cycle cardiaque.
Méthodes
Trente enfants sans cardiopathie ont été étudiés prospectivement. L’anneau aortique a été mesuré dans deux plans orthogonaux en systole et diastole en utilisant un logiciel multiplan 3D et comparé au diamètre 2D.
Résultats
L’âge médian était de 11 ± 3,6 ans. La faisabilité de l’écho 3D était de 100%. Le coefficient de variabilité intra- et interobservateur était de 3,5% et 6%. Le diamètre moyen 2D était inférieur au diamètre 3D en systole (1,94 vs 2,01 mm; p = 0,005). Les mesures 2D et 3D étaient bien corrélées ( p < 0,0001). Les diamètres maximal et minimal 3D de l’anneau aortique étaient différents de façon significative en systole et en diastole soulignant l’excentricité de l’anneau aortique ( p < 0,001). Le diamètre moyen de l’anneau aortique n’était pas différent entre la systole et la diastole avec une importante variabilité individuelle au cours du cycle cardiaque.
Conclusion
Cette étude montre la faisabilité et la reproductibilité de l’écho 3D dans la mesure de l’anneau aortique chez l’enfant. Dans la mesure où l’écho 2D sous-estime le diamètre de l’anneau aortique à cause de son asymétrie, l’écho 3D peut être un examen important avant un geste percutanée ou chirurgicale de la valve aortique.
Background
Aortic valvular stenosis is a relatively common congenital heart disease, involving about 6–7% of infants born with congenital heart disease . Invasive treatment, by either surgical or percutaneous procedures, requires an accurate assessment of the aortic root geometry, especially of the aortic annulus diameter, to minimize the risk of complications, such as aortic regurgitation. Previous studies performed in adults with severe aortic stenosis suggested an underestimation of the aortic annulus diameter with two-dimensional (2D) echocardiography compared with tomography . Only a few studies have addressed this subject in a paediatric population . Real-time three-dimensional echocardiography (3D-TTE) is an emergent non-invasive technique, useful in the evaluation of cardiac chamber volumes and mass, and left ventricular wall motion and in the analysis of morphology and function of heart valves . The aim of our study, therefore, was to investigate the value of 3D echocardiography by comparing 2D-TTE and 3D-TTE measurements of the aortic annulus diameter in children without any cardiac condition.
Methods
Population
Echocardiography data were collected prospectively. Only children with normal cardiac anatomy and function assessed by physical examination, electrocardiography and standard 2D echocardiography were included in the study; real-time 3D-TTE images were then acquired. Children with bicuspid aortic valve or any aortic root disease were not selected. Informed verbal consent was obtained from each patient and legal representatives after a full explanation of the procedure had been given. A written consent form was not required according to French law, given that the echocardiography evaluation was part of the regular management of the children and was required by their medical condition. The study protocol was approved by the National Commission for Data Processing and Freedoms (No. 1673449). No additional examination was performed for the sole purpose of the study. Thirty patients were included between December 2011 and April 2012 in the echocardiography laboratory of the Paediatric Cardiology Unit, Children’s Hospital, Toulouse, France.
Echocardiography and data analysis
Real-time 2D and 3D-TTE were performed using a commercially available IE33 ultrasound system (Philips Medical Systems, Andover, MA, USA). X7-2, X5-1 or X3-1 matrix probes were used, depending on the age of patient. The X7-2 matrix array transducer is particularly well suited to 3D echocardiography in small children .
The examination was performed with the child in a decubitus position. Parameters of gain and compression were optimized. 2D-TTE images were acquired in parasternal long-axis view focused on the aortic annulus area. Aortic annulus diameter was measured in a single plane in end-diastole and mesosystole, between the insertion of the right coronary and non-coronary cusps, using a zoom mode based on international guidelines ( Fig. 1 ). Real-time 3D-TTE images were acquired in parasternal long-axis view using ‘real-time’ 3D mode The 3D loop was turned 90° to obtain a view of the aortic annulus from the left ventricle, to check that the 3D dataset encompassed the whole aortic annulus. The 3D dataset was stored in a DICOM format and transferred to a separate workstation for off-line data analysis. Off-line data analysis was performed using QLab 9 software (Philips Medical Systems, Andover, MA, USA).
Off-line image analysis
Parameters of gain and compression were optimized. Multiplanar reconstruction mode was used, with three independent orthogonal cutting planes. Two orthogonal planes were in the long axis of the left ventricle. The third plane was perpendicular to the two others and moved at the insertion of aortic cusps to obtain a short-axis 2D plane of the aortic annulus. Care was taken to place the plane exactly at the insertion of the leaflets, as recommended . Horizontal and vertical diameters of the aortic annulus were measured in both end-diastole and mesosystole. When the orthogonal diameters were different, the largest diameter was called the maximum diameter (Dmax) and the smallest the minimum diameter (Dmin) (minor and major axes) ( Fig. 2 ).
Reproducibility
Real-time 3D-TTE loop acquisitions were performed by a single confirmed operator. To assess interobserver variability, the 3D dataset was analysed off-line by a second independent operator with the same level of experience of QLab software. To evaluate intraobserver variability, the first operator did the measurements twice, on different days, blinded to the previous results. Variability was not estimated for the 2D measurements.
Statistical analysis
Baseline characteristics were summarized using means and standard deviations (SDs) for continuous variables, and numbers and percentages for categorical variables. First, Spearman’s correlation coefficients were estimated to assess the correlation between 2D and 3D measurements (3D horizontal diameter, 3D vertical diameter, 3D minimum diameter and 3D maximum diameter) with their 95% confidence intervals (CIs) in systole and diastole. A paired Student’s t test was used to compare the mean difference between the 2D and 3D measurements after the assumptions of normality of the dependent variable and homogeneity of variance were checked. Second, an eccentricity index was calculated, expressed as the difference between the maximum and minimum 3D diameters divided by the maximum diameter as a percentage (Dmax–Dmin/Dmax), to assess the aortic annular geometry in systole and diastole. Therefore, an eccentricity index of zero represents a perfect circle, while a progressively higher eccentricity index represents a more elliptical geometry. The Bland-Altman method was used to further investigate the differences between 2D and 3D measurements in systole and diastole . Third, comparisons between systolic and diastolic 2D and 3D aortic annulus diameters were made using a paired Student’s t test, after the assumptions of normality of the dependent variable and homogeneity of variance were checked. The Bland-Altman method was used to further investigate the differences between systolic and diastolic measurements. Finally, interobserver and intraobserver variabilities were investigated by calculating an intraclass correlation coefficient and a variation coefficient. The latter was calculated as the absolute difference between two measurements divided by the average of the two measurements as a percentage. The variation coefficient was calculated as the mean of the variation coefficient for the horizontal and vertical diameters in diastole and in systole. Statistical difference was considered as significant when the p value was < 0.05. Statistical analysis was performed using Stata ® 11.2 software (StataCorp LP, College Station, TX, USA).
Results
Population
Children were referred to the echocardiography laboratory for investigation of a cardiac murmur or before anticancerous chemotherapy. Mean age was 11 ± 3.6 years (range, 3.3–17.6 years); mean height was 142.6 ± 22.2 cm; mean weight was 37.8 ± 14.8 kg (range, 14.2–64.0 kg); mean left ventricular shortening fraction was 35.4 ± 3.5%; and mean end-diastole left ventricular diameter was 36.5 ± 7.1 mm/m 2 .
Feasibility and reproducibility
Aortic annulus diameter measurement was feasible in 2D and 3D in all 30 patients (100%, 95% CI 95.4–100%). The coefficients of variation for intraobserver variability were 3.3% (95% CI 1.7–4.9%) and 5.4% (95% CI 2.3–8.5%) for the horizontal and vertical diameters, respectively. Interobserver variability was 6.8% (95% CI 3.4–10.1%) and 5.3% (95% CI 2.3–8.2%) for the horizontal and vertical diameters, respectively. The intraclass correlation coefficient was 0.91.
Comparison between 2D and 3D aortic annulus diameters in systole
The 2D and 3D aortic annulus diameter measurements in systole are reported in Table 1 . Comparisons between aortic annulus diameter measurements in systole are summarized in Table 2 , as well as Spearman’s correlation coefficients within 2D and 3D measurements. Correlations between the aortic annulus diameter in 2D and the maximum and minimum diameters in 3D were excellent ( r = 0.92, p < 0.0001 and r = 0.88, p < 0.0001, respectively). The correlation coefficients between 2D and horizontal or vertical diameters were r = 0.87 ( p < 0.0001) and r = 0.90 ( p < 0.0001), respectively. However, the 2D aortic annulus diameter in systole was significantly higher than the minimum diameter obtained in 3D ( p = 0.02) and significantly smaller ( p = 0.005) than the maximum diameter obtained in 3D. The mean difference between the 2D diameter and the maximum 3D diameter was 1.1 ± 0.8 mm (95% CI 0.8–1.4 mm). The mean difference between the maximum and the minimum diameters of the aortic annulus was 1.3 ± 0.9 mm (95% CI 0.9–1.6 mm; p < 0.001). The mean eccentricity index in systole was 6.5 ± 4.1% (95% CI 4.9–8.1%). However, given that the maximum aortic annulus diameter was not always in the same axis, there was no significant difference between the mean vertical and horizontal diameters ( p = 0.2). This fact is illustrated by the Bland-Altman analysis ( Fig. 3 A ).
| Aortic annulus diameter | Mean (cm) | 95% CI | Minimum (cm) | Maximum (cm) |
|---|---|---|---|---|
| Systole | ||||
| 2D | 1.94 | 1.83–2.06 | 1.44 | 2.39 |
| 3D vertical | 1.96 | 1.84–2.08 | 1.33 | 2.5 |
| 3D horizontal | 1.93 | 1.80–2.05 | 1.39 | 2.39 |
| 3D minimum | 1.88 | 1.76–2 | 1.33 | 2.39 |
| 3D maximum | 2.01 | 1.89–2.12 | 1.39 | 2.5 |
| 3D mean | 1.95 | 1.83–2.06 | 1.36 | 2.44 |
| Diastole | ||||
| 2D | 1.95 | 1.86–2.04 | 1.5 | 2.4 |
| 3D vertical | 1.95 | 1.85–2.06 | 1.43 | 2.47 |
| 3D horizontal | 1.99 | 1.88–2.1 | 1.46 | 2.62 |
| 3D minimum | 1.93 | 1.82–2.03 | 1.43 | 2.47 |
| 3D maximum | 2.01 | 1.90–2.12 | 1.47 | 2.62 |
| 3D mean | 1.97 | 1.86–2.08 | 1.45 | 2.54 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
