The aim of this study was to demonstrate the feasibility and usefulness of addition of the right parasternal approach to the conventional left parasternal and apical approaches using two-dimensional (2D) and three-dimensional (3D) transthoracic echocardiography (TTE) for morphologic evaluation in cases of transcatheter closure of atrial septal defects (ASDs).
In 112 consecutive patients with ASDs, the morphology of the defects was evaluated for transcatheter closure in the right parasternal view in addition to the conventional left views using 2D and 3D TTE. Measurements of the maximal ASD diameter and detection of deficient rim obtained on 2D TTE were compared with those obtained by 2D transesophageal echocardiography. The shapes and locations of ASDs visualized by 3D TTE were compared with those visualized by 3D transesophageal echocardiography.
In 88 patients (80.0%), optimal images from the right parasternal approach for morphologic evaluation of ASDs were obtained. Although there was a significant difference in maximal ASD diameter obtained only in the conventional left approach compared with transesophageal echocardiographic measurements ( P < .05), when the right parasternal approach was applied, a significant difference was not found ( P = .18), and the diagnostic concordance of the rim deficiency was improved from 85.2% to 90.9%. Three-dimensional TTE from the right parasternal approach improved visualization of the shape and location of ASDs from 65.5% to 74.5%.
Additional use of the right parasternal approach enables detailed morphologic evaluation for transcatheter closure of ASDs. In patients with suboptimal images on 3D TTE in the left conventional approach, additional 3D TTE in the right parasternal approach can improve the feasibility of obtaining optimal 3D images to evaluate the shapes and locations of ASDs.
Transcatheter closure of atrial septal defects (ASDs) has recently become established as a safe and effective treatment, and the procedure has become an alternative to a surgical approach. Appropriate patient selection for transcatheter closure is the most important factor for success in this procedure, and morphologic evaluation, including evaluation of maximal ASD diameter and surrounding rims by echocardiography, is essential. Although two-dimensional (2D) transthoracic echocardiography (TTE) in the left parasternal, apical, and subcostal views is routinely used for this purpose, previous studies have demonstrated that these views enable only limited morphologic evaluation of ASDs. Real-time three-dimensional (3D) echocardiography, in which a comprehensible en face view of ASDs is obtained, has been available in a clinical setting. Three-dimensional TTE is expected to improve understanding of the morphology of ASDs, but data are limited, and it is difficult to obtain good-quality images on 3D TTE using the left parasternal and apical approaches. In this regard, 2D transesophageal echocardiography (TEE) and 3D TEE have been widely accepted and established as diagnostic modalities in evaluation of the morphology of ASDs for transcatheter closure because of their high-quality imaging ; however, TEE has a semi-invasive nature.
The right parasternal approach, in which the transducer is placed to the right of the sternum in the right lateral decubitus position, was reported to enable better visualization of ASDs and evaluation of the direction of shunt flow in patients with ASDs because it obtains a longitudinal vena cava superior-inferior plane of the interatrial septum. In addition, 3D TTE in this approach might improve the feasibility of obtaining optimal en face images of ASDs. However, there have been limited data on the usefulness of the right parasternal approach using 2D and 3D TTE for morphologic evaluation in cases of transcatheter closure. Therefore, we sought to assess the usefulness of the right parasternal approach in addition to the conventional left parasternal and apical approaches in evaluating ASD morphology for the suitability of transcatheter closure using 2D and 3D TTE.
A total of 112 consecutive patients (40 men and 72 women) were prospectively evaluated for transcatheter closure of ASDs using the Amplatzer Septal Occluder (AGA Medical Corporation, Plymouth, MN) with 2D and 3D TTE. Two patients with ASDs other than the secundum type were excluded from this study (one had a superior sinus venosus ASD and the other had an unroofed coronary sinus ASD). Therefore, 110 patients were included in the study. All patients except for one were referred from other hospitals to our institution for transcatheter ASD closure. Age at the examination ranged from 6 to 84 years (mean, 46.1 ± 20.5 years). Two-dimensional TEE and 3D TEE were performed <3 days after TTE by a blinded observer. The study was approved by the local ethics committee.
Two-dimensional TTE was performed using a commercially available ultrasound system with a 3.5-MHz transducer (Vivid 7; GE Healthcare, Wauwatosa, WI). Right ventricular midcavity diameter was measured in the apical four-chamber view according to the guideline of American Society of Echocardiography. In all patients, the morphology of ASDs was evaluated using TTE in the left lateral decubitus position from the left parasternal and apical approaches (conventional left approach). Then a transducer was positioned on the right parasternal border with the patient in the right lateral decubitus position (right parasternal approach). Maximal ASD diameter and the minimal diameter of surrounding rims were measured at end-systole by carefully sweeping the transducer from right to left and top to bottom of the interatrial septum in both approaches. Regarding the maximal ASD diameter, first, the ASD diameter was measured using the conventional left approach (ASD L diameter), and then the ASD diameter was measured by the right parasternal approach (ASD R diameter). The maximal ASD diameter was considered the maximal value from measurements by both approaches. The surrounding rims were classified according to location as superoanterior, inferoanterior, superoposterior, or inferoposterior. The superoanterior rim was measured as the distance between the aorta and the defect. The inferoanterior rim was measured as the distance from the atrioventricular valves. The inferoposterior rim was measured as the distance from the left atrial wall. The superoposterior rim was measured as the distance from the defect to the superior vena cava and to determine the inferoposterior rim as the distance from the defect to the inferior vena cava ( Figure 1 ). Any rim length < 5 mm was considered deficient. First, the presence or absence of a deficient rim was evaluated using the conventional left approach, and then the right parasternal approach was used.
Three-dimensional TTE was performed after 2D TTE using a commercially available ultrasound system with a 3V transducer (Vivid 7). In all patients, the left parasternal approach was first chosen and optimized, and then loops from five consecutive cycles were acquired and digitally stored. In cases with suboptimal 3D images by the left parasternal approach, we attempted to obtain optimal 3D images using the right parasternal approach. In all patients, at least three acquisitions were performed, and the data set with the best image quality was chosen for analysis. The shapes and locations of ASDs were visually evaluated on the best 3D images.
Two-Dimensional and 3D TEE
Two-dimensional and 3D TEE were performed using a commercially available ultrasound system (iE33; Philips Medical Systems, Andover, MA). Maximal ASD diameter (ASD TEE diameter) and minimal diameter of the surrounding rims were assessed at end-systole using both 2D TEE and 3D TEE, as previously reported. To evaluate surrounding rims using 2D TEE, the superoanterior rim was measured as the distance between the aortic annulus and the defect in the horizontal plane at 0° to 30°. The inferoanterior rim was measured as the distance between the defect and atrioventricular valves in the four-chamber view at 135°. The longitudinal plane around 90° was used to determine the superoposterior rim as the distance from the defect to the superior vena cava and to determine the inferoposterior rim as the distance from the defect to the inferior vena cava ( Figure 1 ). The rim length was considered deficient if the length was <5 mm.
Real-time 3D transesophageal echocardiographic data were obtained after a complete 2D transesophageal echocardiographic study. Real-time 3D zoom mode, which displays a smaller, magnified pyramidal data set, was used to evaluate the shapes and locations of ASDs as well as the rough relation to surrounding structures.
Two-dimensional and 3D transesophageal echocardiographic data were considered reference standards. In patients with optimal images obtained on both approaches, ASD TEE diameter and detection of deficient rims obtained on 2D TEE were compared with those obtained on 2D TTE. The shapes and locations of ASDs using 3D TTE were compared with those obtained using 2D and 3D TEE.
ASD diameter and the minimal diameter of the surrounding rims obtained using TTE were measured by two independent observers and by one observer two times 1 month apart in 10 randomly selected patients to determine interobserver variability and intraobserver variability. Variability was assessed as the absolute difference between two measurements expressed as a percentage of their mean values.
Categorical data are expressed as numbers and percentages and continuous data as mean ± SD. The significance of baseline differences was determined using paired and unpaired t tests as appropriate. Categorical variables are expressed as counts and percentages and were compared using χ 2 or Fisher’s exact tests as appropriate. Comparisons between measurements were done using Pearson’s linear regressions analysis. The agreement of the two methods was evaluated using the Bland-Altman test. P values < .05 were considered statistically significant. Statistical analyses were done using SPSS version 18.0 (SPSS, Inc., Chicago, IL).
Baseline Characteristics of Study Population
Table 1 shows the baseline characteristics and 2D transthoracic echocardiographic parameters of the study population. All patients showed hemodynamically significant atrial shunts or the presence of right atrial and ventricular volume overload.
|Age (y)||46.1 ± 20.5 (6–84)|
|Height (m)||1.58 ± 0.12 (1.13–1.83)|
|Weight (kg)||54 ± 12.3 (17–92)|
|Body surface area (m 2 )||1.53 ± 0.22 (0.75–2.14)|
|Right ventricular midcavity diameter (mm)||41.8 ± 5.4 (28–55)|
|Pulmonary flow/systemic flow ratio||2.4 ± 0.7 (1.2–4.1)|
Feasibility of 2D TTE in the Right Parasternal Approach
Two-dimensional TTE with the conventional left approach enabled the detection of shunt flow on color-flow Doppler imaging and visualization of the optimal images for the measurement of defects in all patients. Detection of shunt flow in the right parasternal approach on color-flow Doppler images was successful in 102 patients (92.7%). Optimal images with the right parasternal approach for measurements of defects and surrounding rims were visualized in 88 patients (80.0%). When all patients were divided into two groups according to age, <40 years ( n = 42; mean age, 24.1 ± 10.4 years) and ≥40 years ( n = 68; mean age, 59.6 ± 11.5 years), the percentage of patients in whom optimal images to measure ASD diameter and surrounding rim were obtained in the right parasternal approach was significantly higher in those aged <40 years than in those aged ≥40 years (90.5% vs 73.5%, P = .033).
Data for the 88 patients in whom optimal 2D transthoracic echocardiographic images for measurements of ASD diameter and surrounding rims were obtained by both approaches were analyzed in our study.
Morphologic Evaluation with 2D TTE and TEE
Maximal ASD diameters between ASD L diameter, maximal ASD diameter, and ASD TEE diameter were compared in 88 patients with optimal images from the right parasternal approach. There was a small but significant difference between ASD L diameter and ASD TEE diameter (18.5 ± 6.9 vs 19.0 ± 6.9 mm, P < .05). However, when the diameter obtained with the right parasternal approach was taken into account in addition to the diameter obtained with the conventional left approach, a significant difference was not found between measurements of maximal ASD diameter and ASD TEE diameter (18.8 ± 6.7 mm, P = .18; Figure 2 ). Bland-Altman analysis showed the smallest mean absolute differences and narrower limits of agreement when the measurement from the right parasternal approach was added to that from the conventional left approach ( Figure 3 ).