Closure of Atrial Septal Defects With the Amplatzer Occlusion Device: Preliminary Results

1.1 Device and Delivery System

The Amplatzer and delivery system have been described in detail in previous reports [13]. In brief, this device consists of two self-expandable round disks made of 0.004- to 0.005-in. Nitinol wire mesh that are linked together by a short (4 mm) connecting waist, corresponding to the thickness of the atrial septum (Fig. 1). The left atrial disk extends 7 mm radially around the connecting waist and the right disk 5 mm. The left disk is slightly larger than the right because of the higher left atrial pressure. Both disks are angled slightly toward each other to ensure firm contact with the atrial septum. The prosthesis is filled with Dacron fabric to facilitate thrombosis. Prostheses are currently available in sizes ranging from 4 to 20 mm at increments of 1 mm. The device is attached by a microscrew mechanism onto a 0.038-in. delivery cable made of stainless steel (Fig. 2A). It is loaded into a long 6F or 7F (occluders >10 mm) introducer sheath. For introduction into the delivery sheath the device is pulled into a loader (Fig. 2B).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1

Amplatzer septal occluder made of 0.005-in. Nitinol wire tightly woven into two round disks with a 4-mm connecting waist (arrowheads). Arrowindicates the negative microscrew adaptor mounted on the right atrial disk.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2

A, Delivery cable of the Amplatzer prosthesis (arrowheads). Vertical arrowindicates the plastic vice that facilitates release (unscrewing) of the device. B,Adaptor tube (arrowheads)used for introduction of the device into the delivery sheath.

1.2 Patients

Twenty-two patients from 4 to 14 years of age (median 9.2) with a secundum ASD awaiting surgical closure were evaluated at our institution with two-dimensional and color Doppler echocardiography. Eighteen of these patients met echocardiographic criteria for transcatheter closure with the Amplatzer device (AGA Medical Corporation, Golden Valley, MN) and subsequently underwent device implantation. One of the patients had a significant residual defect after surgical ASD closure and another had severe valvar pulmonary stenosis associated with a secundum ASD.

All devices were implanted under research protocol approved by the Ethical Committee on Clinical Investigation of “Aghia Sophia” Children’s Hospital, Athens. The Amplatzer prosthesis was approved by the Office of Compliance, Center for Devices and Radiological Health of the U.S. Food and Drug Administration for export and investigational use. Informed parental consent was obtained for each patient.

1.3 Study Protocol

Patients were screened by transthoracic two-dimensional Doppler echocardiography with multiple subxyphoid and precordial windows. Five patients with poor transthoracic echocardiographic windows were further evaluated with biplane transesophageal echocardiography. In the last 10 patients transthoracic three-dimensional echocardiography was also performed in order to define the spatial anatomy of the ASDs. Patient inclusion criteria were 1) the presence of an ostium secundum ASD, 2) left to right shunting across the ASD, 3) maximal ASD diameter of <20 mm, 4) a distance of >5 mm from the margins of the defect to the mitral and tricuspid valves, superior vena cava, right upper pulmonary vein and coronary sinus, and 5) dilation of the right ventricle with evidence of right ventricular volume overload. Patients who met these criteria were further evaluated with transesophageal echocardiography under general anesthesia before cardiac catheterization. All patients underwent balloon sizing to establish the stretched diameter of the ASD, which was then used to select the appropriate size of the occluding device. In patients with a balloon-stretched ASD diameter of >21 mm closure was not undertaken. The device size was selected to be equal or 1 mm less than the measured stretched diameter.

1.4 Procedure

The patients were intubated and placed under general anesthesia. After percutaneous puncture of the femoral vein, a complete hemodynamic evaluation was performed with pressure and saturation measurements taken in all cardiac chambers. A pulmonary arteriogram was performed in a 35° left axial oblique projection with 35° cranial angulation (four chamber view) with an 8F Berman angiographic balloon catheter (Arrow International Inc., Reading, PA) to rule out anomalous pulmonary venous connections. Subsequently, contrast medium was injected at the junction of the left atrium with the right upper pulmonary vein to delineate the anatomy of the ASD. A 7F balloon-tipped end-hole catheter (Arrow International Inc., Reading, PA) was manipulated through the ASD into the left upper or lower pulmonary vein. Using an exchange 260 cm, J-tipped guidewire, a 27-mm, 100-cm occlusion balloon catheter (Meditech, Watertown, MA) was introduced into the left atrium. The balloon catheter was inflated with various increments of contrast medium and pulled across the ASD under fluoroscopic and transesophageal echocardiographic observation. A slight deformity of the sizing balloon was used to determine the stretched diameter. The sizing balloon was then removed, reinflated with the same amount of contrast medium and passed through calibrated openings in a sizing plate to determine the stretched diameter. The occluding device was selected to be the same size or 1 mm smaller than the stretched diameter. The device size refers to the diameter of the connecting waist.

The device was screwed to the tip of the delivery cable, immersed in normal saline and drawn into the loader. A Yconnector was applied to the proximal end of the loader to allow flushing with saline. A 7F, long guiding sheath and dilator were advanced over the guidewire through the communication into the left atrium. The correct position of the delivery sheath was verified by a test injection of contrast medium. The loader with the collapsed device was then advanced into the guiding catheter by pushing the delivery cable (Fig. 3A). Under fluoroscopic (Fig. 3B) and ultrasonic guidance, the left atrial disk was deployed and pulled gently against the atrial septum, which was both felt and observed by transesophageal echocardiography. Using gentle tension on the delivery cable, the sheath was pulled back and the right atrial disk was deployed (Fig. 3C). To and fro motion of the delivery cable ensured a secure position across the ASD. This was also observed by two-dimensional Doppler echocardiography and fluoroscopy. Biplane transesophageal color Doppler echocardiography was performed after device deployment to check for the presence of residual shunt, possible obstruction to systemic or pulmonary venous return and impairment of the atrioventricular valves. Once its position was optimal, the device was released by counterclockwise rotation of the delivery cable. After release of the Amplatzer both color Doppler echocardiography and pulmonary angiography were performed to detect any residual shunt (Fig. 3DFig. 4). In 10 patients accurate device position was confirmed by en-face three-dimensional echocardiographic views from the right and left atrium. Intravenous antibiotics (ampicillin 200 mg/kg body weight and gentamycin 5 to 6 mg/kg) were delivered immediately after placement of the prosthesis and repeated at 8 and 16 h, for a total of three doses. All patients were discharged on the day after the procedure with instructions to take aspirin 3 to 5 mg/kg daily for 3 months.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 3

Cineradiographic frames (A to C)and angiogram (D)in four-chamber view showing steps of atrial septal defect closure with the Amplatzer: A,Advancement of the device through a 7F delivery sheath into the left atrium. B,Deployment of the left atrial disk. C,Deployment of the right atrial disk. D,Levo-phase of pulmonary arteriogram after release of the device, showing complete closure and a good position of both disks across the atrial septum.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 4

Transesophageal two-dimensional and color Doppler images of the atrial septum obtained from short-axis views before (top)and immediately after (bottom)implantation of the Amplatzer prosthesis. Note good position of the device (white arrows)with no evidence of residual shunt by color Doppler. D = diameter of the defect; LA = left atrium; RA = right atrium.

1.5 Follow-Up Studies

A chest radiograph and a transthoracic color Doppler echocardiographic study were performed on all patients at 24 h after the procedure and at 1 and 3 months later. A transesophageal echocardiographic study was scheduled for the 6-month follow-up.

Persistent atrial shunts were angiographically (immediately after the procedure) and echocardiographically defined as “foaming” (minor diffuse leak through the Dacron fabric), trivial (faint right atrial opacification and jet <1 mm in diameter), small (minor right atrial opacification and jet 1 to 2 mm in diameter), moderate (obvious right atrial opacification and jet 3 to 4 mm in diameter) and large (intense right atrial opacification and jet >4 mm in diameter).

1.6 Statistical Analysis

Results are expressed as mean value ± SD, with confidence intervals given where applicable. The Student paired ttest was used to compare transesophageal echocardiographic and balloon-stretched ASD diameter. A p value <0.05 was considered statistically significant.

3.1 Comparison with Other Devices

The Amplatzer was designed to overcome many of the drawbacks of previously used devices. Previous devices include the clamshell septal occluder [3, 4], the Sideris prosthesis (buttoned device) [5, 7], the atrial septal defect occlusion system [8, 9], the Das Angel Wing [10, 11]and the Pavcnik monodisk [12]. Besides the Sideris prosthesis, which is delivered through an 8F sheath for devices up to 40 mm, all other occluders require a large 9F to 13F sheath, which makes their application difficult or impossible in small children. The Amplatzer is delivered through a 6F to 7F sheath, which facilitates its use even in infancy. Moreover, the other devices are limited in that they require a large device size (2- to 2.5-fold the stretched ASD diameter) to ensure complete closure if they are not perfectly centered. The Amplatzer is a self-centering prosthesis and the round retention disks extend radially beyond the defect, resulting in a much smaller overall size than all other devices. The Das Angel Wing and the Pavcnik monodisk are the only currently used devices with a self-centering mechanism [10, 12]. Of these two devices, the Pavcnik monodisk has the advantage of requiring a rather small device/ASD diameter ratio but it offers a questionable centering mechanism and there is very limited clinical experience. The Das Angel Wing prosthesis has a more reliable centering mechanism, but not during deployment, and consequently, the retention skirts have to remain oversized.

In contrast to the atrial septal occluders that accomplish closure of the ASD by the retention flanges (patches), the Amplatzer uses a short (4 mm) communicating waist to stent the defect, forcing blood flow through a network of highly thrombogenic polyester material. Indeed, we observed “foaming” with transesophageal echocardiography and angiography immediately after deployment, which completely disappeared within 15 to 20 min. Moreover, the inward inclination of both retention disks allows firm contact with the atrial septum, which enhances endothelialization and reduces the risk of residual shunting. The previously reported incidence of residual shunting (20% to 39%) is much higher than in the present study (6%) [4, 7, 9, 11].

Selection of the appropriate device size is of paramount importance for effective and safe ASD closure regardless of device type. Balloon sizing to establish the stretched ASD diameter has remained the gold standard in spite of the more convenient echocardiographic measurement. This technique is less reliable because many communications are not perfectly round [4, 14]. Three patients in our study had a trivial residual shunt that disappeared in two patients within 1 month and 3 months, respectively, after the procedure. It should be noted that in all three patients the size of the device was 1 mm less than the measured balloon-stretched ASD diameter. Therefore, we believe that it is essential to choose a device identical to the measured balloon-stretched diameter of the defect in order to maximize the efficacy of this particular design.

A very important property of the Amplatzer is retrievability while the device is still attached to the delivery cable, which allows repositioning in case of misplacement, thus obviating surgical removal. Most of the other devices, with the exception of the clamshell septal occluder [4], are not retrievable or once deployed it is very laborious to reposition or to remove them [7–9, 11, 12]. The Amplatzer prosthesis can be fully recaptured into the sheath with the delivery cable. Because of its superelastic properties, the nitinol frame of the device retains its initial shape even if it is deployed several times. Indeed misplacement of both discs across the defect was encountered in one of our patients and was easily corrected by withdrawing and redeploying the device. An additional advantage of this prosthesis are the two round disks. Devices with corners or spokes may cause accidental injury of atrioventricular valves or perforations of an intracardiac structure during disk deployment. Also, delayed perforations can occur. Furthermore, complications previously reported, such as metal fatigue fractures [15], structural failure or embolizations of the device [4–7, 16], atrial perforation [7, 16]and atrioventricular valve damage [7], were not observed in the present study, although the follow-up is short.

As compared with other devices, the Amplatzer’s loading, technical deployment and recapturing is simple, without complicated mechanisms. This significantly reduces the procedural and fluoroscopy time and shortens the learning curve of each operator. Indeed the procedural (67.5 ± 8.1 min) and fluoroscopy (20.3 ± 3.5 min) times in this study were very short for this type of intervention. In the atrial septal defect occlusion system experience in children, procedural times ranged from 132 to 194 min and fluoroscopy times from 17 to 39 min [9].

Biplane transesophageal color Doppler echocardiography was used in this study for guidance of transcatheter closure of the ASDs. Echocardiographic imaging is superior to fluoroscopy as it provides unique information regarding the anatomy of the defect, the position of the device across the atrial septum and its relation to adjacent cardiac structures [17, 18]. An important benefit of the Amplatzer is its optimal echogenicity, owing to its Nitinol mesh structure, which makes it easily visible in the short-axis and longitudinal views of the device and residual shunting can be precisely delineated.

Recently, three-dimensional transesophageal or transthoracic echocardiography was used for patient selection and guidance of transcatheter closure of ASDs [19, 20]. Three-dimensional echocardiography displays the defect as a dynamic structure whose size and morphology change with the cardiac cycle, and its maximal diameter can be better appreciated when the periphery of the defect is seen in an en-face view. Additionally, one can visualize both atrial discs of the occlusion device and their dynamic anatomic relation to the adjacent cardiac structures. In this study the use of three-dimensional transthoracic echocardiography proved useful in the preselection of patients and provided guidance of ASD closure with the Amplatzer prosthesis.

3.2 Conclusions

Our results indicate that the Amplatzer is an efficient prosthesis that can be safely used in children with secundum ASDs. It is simple in construction, easy to deploy and can be withdrawn and repositioned many times. Further studies are required to establish its value in a larger number of patients, for closure of larger ASDs and patent foramen ovale. Furthermore, long-term follow-up is needed to exclude possible adverse side effects.