Article Text
Abstract
OBJECTIVE To apply three dimensional echocardiography to describe the geometric profile of the Amplatzer and Cardioseal occluders after deployment for closure of atrial septal defect.
METHODS 20 patients (mean (SD) age, 14 (5) years) were enrolled for transcatheter closure of a secundum atrial septal defect with the Amplatzer occluder (10) or with the Cardioseal occluder (10). The two populations were matched for the stretched diameter of the defect (mean 18 (6) mm). The profile of the two occluders was examined.
RESULTS Transoesophageal echocardiography did not show any residual shunts after Amplatzer occluder deployment, whereas three patients had a small residual leak after Cardioseal deployment. One patient had transient atrioventricular block with the Amplatzer device. The mean surface area of the Amplatzer occluder was 6.9 (2) cm2, and that of the Cardioseal device 5.4 (3) cm2 (p = 0.03). The mean volume of the Amplatzer occluder was 9.2 (1) cm3, while that of the Cardioseal occluder was 3.5 (1) cm3 (p < 0.0001). From the three dimensional views, the Cardioseal occluder looked like a flat square after deployment whereas the Amplatzer occluder took up a ball shape in the atrial cavity.
CONCLUSIONS Three dimensional views by multiplane transoesophageal echocardiography allow a realistic in vivo description of atrial septal occluders. The Amplatzer occluder, with its high geometric profile, allows complete closure of large atrial septal defects but with some risk of mechanical complications. Use of the Cardioseal device, with its small surface coverage and high residual shunt rate, should be limited to transcatheter closure of a patent foramen ovale or small atrial septal defects.
- atrial septal occluder
- three dimensional echocardiography
Statistics from Altmetric.com
Although the low surgical mortality and morbidity associated with closure of atrial septal defects is generally accepted, transcatheter closure is becoming an attractive alternative, because it eliminates or reduces the risks and morbidity inherent in surgery. Several transcatheter techniques and devices have been developed for transcatheter occlusion of defects. In particular, the Amplatzer and Cardioseal occluders are commonly used for closing atrial septal defects, with a high success rate (fig 1).1 ,2 However, questions have been raised about the profiles of these devices, and mechanical complications have been reported after deployment of septal occluders.3
While it is difficult to visualise the occluder configuration with cross sectional echocardiography, three dimensional echocardiography allows en face views of intracardiac structures such as the atrial septum.4 Volumetric measurements are also possible using the three dimensional method.5 Our aim was to apply three dimensional echocardiography to describe the geometric profile of atrial septal occluders after deployment.
Methods
PATIENTS
Twenty patients referred for transcatheter closure of a secundum atrial septal defect were enrolled. Eight were male and 12 female. Their mean (SD) age was 14 (5) years, range 5–20 years. The atrial septal defect was closed with the Amplatzer occluder in 10 patients and with the Cardioseal occluder in the other 10. The two populations were matched for the stretched defect diameter, measured by the balloon transcatheter method.
THREE DIMENSIONAL ECHOCARDIOGRAPHIC ACQUISITION
Transoesophageal echocardiography with three dimensional reconstruction was performed under general anaesthesia after deployment of the septal occluder. The system consisted of a 5–7 Hz multiplane transoesophageal probe connected to an Acuson Sequoia C256 platform (Acuson, Mountain View, California, USA). This was linked to a three dimensional reconstruction system (Echo-scan, TomTec GmbH, Munich, Germany). Multiplane transoesophageal echocardiography and three dimensional echocardiography were performed as previously described.6 The transoesophageal probe rotation was commanded by a steering logic for controlled image acquisition.
Electrocardiography and respiratory gating by thoracic impedance measurements ensured optimal spatial and temporal registration of the cardiac images. RR intervals were predetermined with a variability of 100 ms or less. Respiration was gated at the end expiratory phases.
The reference position was a basal transverse plane to visualise the long axis of the septal occluder. From this reference image, 90 sequential cross sections acquired every 2° were obtained, each during a complete cardiac cycle, and stored in the computer memory. On average, two acquisitions were obtained in each patient and the one with the least motion artefact was chosen for three dimensional measurements. The recorded images were formatted in their correct rotational sequence according to their ECG phase in volumetric datasets. Postprocessing of the datasets was performed off line, using the system's analysis program (Echo-view, TomTec). To fill the gaps in the far fields, a trilinear cylindrical interpolation algorithm was used.
DATA ANALYSIS
From the conical data volume, specific cut planes were used for three dimensional reconstructions to image the septal occluder en face from both the right and the left atrium. A transparency was placed over the extracted three dimensional image. The surface area of the septal occluder was traced using an off line measurement system (Freeland cine view, version 4, TomTec). The septal occluder was sliced into multiple equidistant parallel long axis views. The surface area of each cross section was measured by manual endocardial tracing and the volume of each slice calculated. The entire volume of the septal occluder was automatically derived by summing the volumes of each slice.5 All values used for comparison are reported as mean (SD). Unpaired t tests were used to analyse the grouped data. Significance is defined as p < 0.05.
Results
The mean (SD) stretched diameter was 18 (6) mm, range 12–25 mm. The Amplatzer septal occluder sizes were from 12 to 26 (equal to the stretched diameter), while the Cardioseal occluder sizes were from 23 to 40 (stretched diameter × 1.8). Transoesophageal echocardiography did not show any residual shunt after deployment of the Amplatzer occluder, whereas three patients had a small residual leak (< 4 mm) around the Cardioseal device. One patient had a transient atrioventricular block after deployment of a size 26 Amplatzer device. No venous flow or valvar obstruction was observed.
En face three dimensional views of the septal occluder were obtained in all study patients, and it was possible to measure the surface area and the volume of the septal occluder. Acquisition time was within five minutes in all cases. The time needed for three dimensional measurements was 20 minutes.
The mean surface area of the Amplatzer septal occluder was 6.9 (2) cm2 and of the Cardioseal occluder 5.4 (3) cm2 (p = 0.03). The mean volume of the Amplatzer septal occluder was 9.2 (1) cm3, while that of the Cardioseal occluder was 3.5 (1) cm3 (p < 0.0001). From the three dimensional views, the Cardioseal occluder after deployment looked like a flat square (fig 2), while the Amplatzer occluder took up a ball shape in the atrial cavity.
Discussion
This study shows that three dimensional views of atrial septal occluders by multiplane transoesophageal echocardiography provide realistic visualisation of the deployment of the occluder in the atrial cavity. Measurements of the surface area and volume of the septal occluder are possible using the three dimensional method.
This study is the first to report the profile of the Amplatzer occluder from three dimensional views. The volumetric data measured by the extracted three dimensional images confirmed its high geometrical profile. Mechanical complications such as atrioventricular block or superior vena caval obstruction have been reported after deployment of the Amplatzer occluder.3 (In our small series we observed one transient atrioventricular block after deployment of a size 26 Amplatzer device.) These complications could be explained by the relatively large volume of the device, which may compress the atrioventricular node or obstruct venous flow. The mitral valve might also be compromised after deployment of the Amplatzer occluder.7 Fortunately the Amplatzer device was designed to be easily retrievable until released from the delivery wire, thereby allowing removal in the event of malpositioning. The large surface area covered by this device contributed to the very small amount of residual shunting. European series reported occlusion rates of at least 97% at three months.1 ,3 ,7 Despite its high geometric profile, the Amplatzer occluder is the preferred device in many institutions because of its ease of handling. Moreover, it allows closure of large atrial septal defects (over 25 mm diameter).8
Other studies have reported three dimensional reconstructions of septal occluders of the double umbrella type, such as Das-Angel wings and the Cardioseal device.9 ,10 Transcatheter occlusion with the Cardioseal has several difficulties. The delivery technique is complicated and the device is not easily retrievable. Moreover, after deployment the rate of residual shunting is high.2 This may be explained by the small surface area covered by the occluder. The absence of a self centring mechanism also increases the risk of incomplete closure. The new flexible, self centring device (the Starflex occluder) may improve the rate of total occlusion.11 The use of these double umbrella occluders should be limited to the transcatheter closure of patent foramina ovale or small atrial septal defects.
Selection of patients for transcatheter closure of atrial septal defects requires accurate information about the anatomy of the defect. Exact measurements of its maximum diameter and of the size of the circumferential tissue rim allows one to choose the most suitable device for transcatheter closure. On three dimensional echocardiography, secundum atrial septal defects are seen to be complex structures.12 From three dimensional views, variations in shape and changes in the surface area of the defect during the cardiac cycle have been reported.13 ,14 Three dimensional echocardiography has enhanced our understanding of the anatomy and pathophysiology of secundum atrial septal defects. En face three dimensional views can reveal the occluder profile in vivo. However, a decrease in the time required for three dimensional reconstruction and measurement is needed to make clinical application easier.
Several potential problems in the three dimensional reconstructions need to be addressed. From these reconstructions, atrial septal occluders often appeared to be thicker than they actually are, with decreased resolution. This is in part related to a spatial volume artefact and slow frame rates. The ability to make three dimensional reconstructions directly from the digital images without transformation into video signals would improve the quality of the images, which is still very dependent on the gain settings.
Our study is limited by the small number of patients and the range of devices. Further studies, including a larger population, would be helpful to determine whether the three dimensional reconstructions are predictive of device success, complications, or failure in a given patient, in relation to the size of the atrial septal defect or its proximity to surrounding structures. However, three dimensional views by multiplane transoesophageal echocardiography allow a realistic in vivo description of the atrial septal occluder. The Amplatzer occluder, with its high geometric profile, allows complete closure of large atrial septal defects but with some risk of mechanical complications. Use of the Cardioseal device, with its rather small surface coverage and high rate of residual shunting, should be limited to the transcatheter closure of a patent foramen ovale or small atrial septal defects.
In conclusion, three dimensional echocardiography is an important technique in future initiatives to develop transcatheter devices for closing defects.