DOI: 10.1148/rg.23si035509
(Radiographics. 2003;23:S51-S58.)
© RSNA, 2003
PERIOPERATIVE CROSS-SECTIONAL IMAGING |
Evaluation of a Large Atrial Septal Occluder with Cardiac MR Imaging1
Chantale Lapierre, MD,
Marie-Josée Raboisson, MD,
Joaquim Miró, MD,
Nagib Dahdah, MD and
Ronald Guérin, MD
1 From the Departments of Medical Imaging (C.L., R.G.) and Cardiology (M.J.R., J.M., N.D.), Hôpital Sainte-Justine, University of Montreal, 3175 Côte-Sainte-Catherine Road, Montreal, Quebec, Canada H3T 1C5. Presented as a scientific poster at the 2002 RSNA scientific assembly. Received February 13, 2003; revision requested April 22 and received June 17; accepted July 7. Address correspondence to C.L. (e-mail: chantal_lapierre@ssss.gouv.qc.ca).
 |
Abstract
|
|---|
Magnetic resonance (MR) imaging was used to evaluate the position of a large atrial septal occluder (ASO) with regard to adjacent cardiac valves and veins and to assess any negative effects of the ASO on these vital structures. A total of 26 pediatric patients (mean age, 4.5 years; mean interval after implantation, 18.8 months) were evaluated with cardiac MR imaging. The position of the ASO was best depicted with two-dimensional cine fast low-angle shot imaging. The authors observed impingement of the ASO on the right superior pulmonary vein in 14 patients, on the right inferior pulmonary vein in three patients, on the right superior vena cava in 13 patients, and on the right inferior vena cava in nine patients. In two patients, protrusion of the ASO into the right inferior vena cava was associated with coronary sinus prominence. The ASO was in contact with the mitral valve in 10 patients. Deformation of the aortic valve and root was evident at the onset of the R wave in 19 patients and persisted throughout the cardiac cycle in five of these patients. Cardiac MR imaging reliably depicted the position of the ASO with regard to vital structures. Despite the protrusion of the ASO, no significant effects on venous structures or cardiac valves were observed.
© RSNA, 2003
Index Terms: Atrial septal defect, 514.14 • Heart, MR, 50.12141 • Heart, prostheses, 514.452 • Magnetic resonance (MR), cine study
|
Introduction
|
|---|
Transcatheter closure of secundum atrial septal defects was first described by King et al (1) more than two decades ago. Since then, the procedure has been used with increasing frequency as an alternative to open-heart surgery. The Amplatzer atrial septal occluder (ASO) (AGA Medical, Golden Valley, Minn) is the first choice in most leading institutions for the closure of atrial septal defects in children (Fig 1). To our knowledge, only one occurrence of a life-threatening complication (perforation of the aorta in an adult patient) related to implantation of this prosthetic device has been reported in the literature (2). Although most pediatric cardiologists currently use this ASO in their patients, some investigators have suggested that the interaction of this relatively large device with adjacent native structures may be problematic in young children (3–6).
View larger version (187K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Photograph of the Amplatzer ASO. The device consists of two saucer-shaped disks made of self-expanding nitinol mesh and connected by a central cylinder that resembles a stent-graft.
|
|
The use of cardiac magnetic resonance (MR) imaging to assess the size and location of atrial septal defects and to guide the real-time positioning of ASOs has been reported previously (7–9). Our aim in this study was to evaluate the capability of MR imaging to accurately depict the position of the ASO with respect to adjacent structures (the right pulmonary veins, the superior and inferior venae cavae, the coronary sinus, the mitral valve, and the aortic valve and root) and to assess its effect on those structures.
|
Materials and Methods
|
|---|
Study Population
Between December 1997 and February 2002, 122 patients younger than 12 years of age underwent implantation at our institution of Amplatzer ASOs for closure of atrial septal defects. Since the dimensions and growth of the cardiac structures are proportional to patient height, the ratio of the left atrial disk diameter (mm) to patient height (cm) was plotted against age. Patients with ratios in the upper quartile were included in our study population (n = 29). Two patients refused to undergo MR imaging, and one patient was unavailable for follow-up. The remaining 26 patients were evaluated with cardiac MR imaging (mean age ± standard deviation, 4.5 years ± 2.8; mean interval after implantation, 18.8 months ± 11).
MR Imaging Technique
After the skin was prepared with an abrasive gel, precordial fiber-optic electrodes were attached in the region of the cardiac apex by the MR technologist. An intravenous sedative (7 mg of pentobarbital per kilogram of body weight) was administered to patients under the age of 8 years (n = 9) before MR imaging. Arterial blood oxygen saturation was monitored with a pulse oximeter, and particular attention was paid to electrocardiographic tracings during MR image acquisition. In all patients, MR imaging was performed with a 1.5-T unit (Symphony; Siemens Medical Systems, Erlangen, Germany) and a flexible coil. The total duration of the examination was approximately 60 minutes. All 26 patients were imaged by the same radiologist (C.L.).
Morphologic Examination.—
Axial single-section images of the whole heart, from the vascular pedicle to the diaphragm, were obtained with a T1-weighted turbo spin-echo sequence while the patient breathed normally. The following parameters were used: 320–980/15–30 (repetition time msec/echo time msec); matrix, 160 x 256; maximum field of view, 380–440 mm; turbo factor, 3; three or four signals acquired; and imaging time per section, 30–150 seconds.
Dynamic Examination.—
To dynamically assess the relationship between the ASO and adjacent vital structures during the cardiac cycle, two-dimensional (2D) cine MR imaging was performed with a fast low-angle shot (FLASH) in selected planes and without patient breath holding. For this purpose we have favored the 2D FLASH over true fast imaging with steady-state precession because the former technique is more sensitive to the "jet phenomenon" and provides better depiction of the valvular folds. The planes of imaging were selected on the basis of T1-weighted MR images that had been obtained previously. The goal was to achieve depiction both of the venous openings within the atria and of the ASO. For imaging of the venae cavae, coronal oblique sections were optimal. In imaging of the pulmonary veins, the sequence was applied twice: first in the coronal oblique plane along the axis of the veins, and then in the axial oblique plane along the same axis. Imaging of the coronary sinus also was performed in two planes: the coronal oblique plane parallel to the axis of the atrioventricular groove, and the plane perpendicular to that axis. Imaging of the aortic valves was performed in the coronal plane for localization, and then in the axial oblique plane parallel to the axis of the valve. Finally, the mitral valve was imaged to obtain a four-chamber view. The following parameters were used: 49/4.6; flip angle, 20°; matrix, 208 x 256; maximum field of view, 380–440 mm; two or three signals acquired; and imaging time per section, 30–75 seconds.
Image Analysis
MR images were examined for evidence of protrusion (ie, overhanging) of one or both disks of the ASO into the opening of the pulmonary or systemic veins in the atria during the cardiac cycle. When protrusion was observed, the extent of protrusion was calculated as a ratio of the length of the disk protrusion to the total diameter of the venous structure involved (Fig 2). Images on which protrusion was observed were further examined for any evidence of dilatation in the corresponding draining vein or of the jet phenomenon in the valves or venous openings. Special attention was given to ruling out a left superior vena cava in patients in whom a prominent coronary sinus was observed. Observations of venous dilatation—that is, of any increase in luminal caliber that might indicate obstruction—were likewise noted. Images also were evaluated for signs of contact between the ASO and the mitral valve leaflet or annulus, extrinsic compression and deformation of the aortic valve and root by the ASO, and valvular regurgitation.
View larger version (133K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. The percentage of protrusion was calculated from the ratio of the extent of disk protrusion (solid arrows) to the total diameter of the adjacent structure (in this case, the superior vena cava) (open arrows).
|
|
|
Results
|
|---|
Protrusion of the ASO into openings of the pulmonary and systemic venous returns was observed on MR images (Table) at various points in the cardiac cycle. With regard to the right pulmonary venous return, protrusion of the ASO occurred mainly in the superior vein when the left atrium was at its minimal volume. The extent of the protrusion was approximately 50%–75% in most cases (Figs 3, 4). With regard to the systemic venous return, protrusion of the ASO occurred into the openings of the venae cavae (Figs 5, 6). Maximum protrusion of the ASO in the inferior vena cava occurred when the left atrium was at its maximal volume, whereas maximum protrusion in the superior vena cava occurred both when the left atrium was at its minimal volume and when it was at its maximal volume. The range of variations in the percentage of protrusion was roughly the same in the superior and inferior venae cavae, but a protrusion of more than 75% into the opening of the inferior vena cava was noted in one patient who presented with no clinical signs of obstruction. No venous dilatation or evidence of the jet phenomenon was noted in any of the structures evaluated.
View larger version (144K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. Coronal oblique 2D cine MR images of the right superior pulmonary vein. (a) Image obtained with the left atrium (straight black arrow) at maximal volume shows that the left disk of the ASO (curved black arrow) does not protrude into the vein (white arrow). (b) Image obtained with the left atrium (straight black arrow) at minimal volume shows protrusion of the disk (curved black arrow) into the vein (white arrow).
|
|
View larger version (140K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. Coronal oblique 2D cine MR images of the right superior pulmonary vein. (a) Image obtained with the left atrium (straight black arrow) at maximal volume shows that the left disk of the ASO (curved black arrow) does not protrude into the vein (white arrow). (b) Image obtained with the left atrium (straight black arrow) at minimal volume shows protrusion of the disk (curved black arrow) into the vein (white arrow).
|
|
View larger version (126K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. Axial oblique 2D cine MR images of the right inferior pulmonary vein. (a) Image obtained with the left atrium (straight white arrow) at maximal volume shows that the left disk of the ASO (black arrow) does not protrude into the vein (curved white arrow). (b) Image obtained with the left atrium (straight white arrow) at minimal volume shows protrusion of the disk (black arrow) into the vein (curved white arrow).
|
|
View larger version (128K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. Axial oblique 2D cine MR images of the right inferior pulmonary vein. (a) Image obtained with the left atrium (straight white arrow) at maximal volume shows that the left disk of the ASO (black arrow) does not protrude into the vein (curved white arrow). (b) Image obtained with the left atrium (straight white arrow) at minimal volume shows protrusion of the disk (black arrow) into the vein (curved white arrow).
|
|
View larger version (143K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5a. Sagittal oblique 2D cine MR images of the right superior vena cava. (a) Image obtained with the left atrium (straight white arrow) at maximal volume shows that the right disk of the ASO (black arrow) does not protrude into the vein (curved white arrow). (b) Image obtained with the left atrium (straight white arrow) at minimal volume shows protrusion of the disk (black arrow) into the vein (curved white arrow).
|
|
View larger version (142K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5b. Sagittal oblique 2D cine MR images of the right superior vena cava. (a) Image obtained with the left atrium (straight white arrow) at maximal volume shows that the right disk of the ASO (black arrow) does not protrude into the vein (curved white arrow). (b) Image obtained with the left atrium (straight white arrow) at minimal volume shows protrusion of the disk (black arrow) into the vein (curved white arrow).
|
|
View larger version (129K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6a. Sagittal oblique 2D cine MR images of the inferior vena cava. (a) Image obtained with the left atrium (straight white arrow) at maximal volume shows the greatest protrusion of the right disk of the ASO (black arrow) into the vein (curved white arrow). (b) Image obtained with the left atrium (straight white arrow) at minimal volume shows less protrusion of the disk (black arrow) into the vein (curved white arrow).
|
|
View larger version (130K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6b. Sagittal oblique 2D cine MR images of the inferior vena cava. (a) Image obtained with the left atrium (straight white arrow) at maximal volume shows the greatest protrusion of the right disk of the ASO (black arrow) into the vein (curved white arrow). (b) Image obtained with the left atrium (straight white arrow) at minimal volume shows less protrusion of the disk (black arrow) into the vein (curved white arrow).
|
|
We were unable to assess satisfactorily the position of the ASO with regard to the opening of the coronary sinus in the right atrium because of the brief duration of sedation and the small size of this structure in pediatric patients. The prominence of the coronary sinus in association with a protrusion into the inferior vena cava and the absence of a left superior vena cava on images of two patients led us to suspect protrusion by the ASO into the opening of the coronary sinus in the right atrium—especially likely, given the low position of the ASO with respect to the interatrial septum. However, these two patients had no clinical symptoms of exertional cardiac ischemia.
Images of 10 patients showed contact between the left disk of the ASO and the mitral valve, with no resultant deformation or regurgitation (Fig 7). The ASO was in contact with the annulus in six patients (Fig 8), the anterior leaflet in three patients, and both the annulus and the anterior leaflet in one patient.
Extrinsic contact of the ASO with the aortic valve and root was observed in all patients (n = 26) (Figs 9, 10), but this contact led to deformation in only 19 patients—by the right disk in three patients, by the left disk in 10 patients (Fig 11), and by both disks in six patients. For the most part, deformation caused by the left disk was noted only when the left atrium was at its maximal volume (in nine of 10 patients), whereas deformation caused by the right disk was noted both when the left atrium was at its minimal volume and when it was at its maximal volume. However, when both disks were involved (in six of six patients), deformation caused by the left disk occurred when the left atrium was at its maximal volume, whereas deformation caused by the right disk occurred when the left atrium was at its minimal volume. Deformation was present throughout the cardiac cycle in five of the 19 patients (Fig 12). In six patients, increased distance between the anterosuperior portions of the two disks was observed near the root and the aortic annulus during the cardiac cycle, accompanied by increased distance between the right disk and the interatrial septum (Fig 13).
View larger version (126K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11. Axial oblique 2D cine MR image shows extrinsic contact between the left disk of the ASO (curved arrow) and the aortic valve (straight arrow) and root, with consequent deformation.
|
|
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 12. Axial oblique 2D cine MR image shows extrinsic contact between both disks of the ASO (curved arrow) and the aortic valve (straight arrow) and root, with consequent deformation.
|
|
View larger version (136K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13a. (a) Axial oblique 2D cine MR image obtained during ventricular systole shows displacement of the right disk (curved arrow) in the right atrium and an increased distance between the disks (straight arrow). (b) Axial oblique 2D cine MR image obtained during ventricular diastole shows less distance between the disks (straight arrow). Curved arrow = right disk.
|
|
View larger version (137K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13b. (a) Axial oblique 2D cine MR image obtained during ventricular systole shows displacement of the right disk (curved arrow) in the right atrium and an increased distance between the disks (straight arrow). (b) Axial oblique 2D cine MR image obtained during ventricular diastole shows less distance between the disks (straight arrow). Curved arrow = right disk.
|
|
|
Discussion
|
|---|
To our knowledge, this study is the first on this topic. Echocardiography frequently is used for direct, noninvasive evaluation of ASOs; however, ultrasound is limited by its inability to penetrate air and bone, and a variable acoustic window may impair the visualization of cardiac structures. Cardiac MR imaging in this context has clear advantages over echocardiography.
At our institution, all cardiac MR examinations are supervised by a radiologist. However, given that most such examinations are fairly easy to perform in a reasonable amount of time, a senior cardiac MR technologist also could fulfill the supervisory role. The greatest difficulties that we have encountered in MR imaging of the ASO have been caused by an abnormal axis of the atrial septum in some patients. The presence of this condition was established by means of chest radiographs obtained prior to MR imaging. Cardiac MR imaging performed in patients who have this condition should be supervised by a radiologist.
Despite the protrusion of ASOs into venous structures and contact between ASOs and aortic or mitral valves, no significant repercussions (eg, pulmonary or systemic venous obstruction, progressive aortic or mitral regurgitation) were observed in this study. However, long-term follow-up will be necessary in order to fully evaluate the effects of large ASOs on these vital structures in pediatric patients.
|
Footnotes
|
|---|
Abbreviations: ASO = atrial septal occluder,
FLASH = fast low-angle shot,
2D = two-dimensional
|
References
|
|---|
- King TD, Thompson SL, Steiner C, Mills NL. Secundum atrial septal defect: nonoperative closure during cardiac catheterization. JAMA 1976; 235:2506-2509.[Abstract]
- Aggoun Y, Gallet B, Acar P, et al. Perforation of the aorta after percutaneous closure of an atrial septal defect with an Amplatz prosthesis, presenting with acute severe hemolysis. Arch Mal Coeur Vaiss 2002; 95:479-482.[Medline]
- Holmvang G, Palacios IF, Vlahakes GJ, et al. Imaging and sizing of atrial septal defects by magnetic resonance. Circulation 1995; 92:3473-3480.[Medline]
- Taylor AM, Stables RH, Poole-Wilson PA, Pennell DJ. Definitive clinical assessment of atrial septal defect by magnetic resonance imaging. J Cardiovasc Magn Reson 1999; 1:43-47.[Medline]
- Holmvang G. A magnetic resonance imaging method for evaluating atrial septal defects. J Cardiovasc Magn Reson 1999; 1:59-64.[Medline]
- Durongpisitkul K, Tang NL, Soongswang J, Laohaprasitiporn D, Nana A, Kangkagate C. Cardiac magnetic resonance imaging of atrial septal defect for transcatheter closure. J Med Assoc Thai 2002; 85(suppl 2):S658-S666.
- Rickers C, Jerosch-Herold M, Hu X, et al. Magnetic resonance image-guided transcatheter closure of atrial septal defects. Circulation 2003; 107:132-138.[CrossRef][Medline]
- Weber C, Dill T, Mommert I, Hofmann T, Adam G. The role of MRI for the evaluation of atrial septal defects before and after percutaneous occlusion with the Amplatzer septal occluder®. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 2002; 174:1387-1394.[Medline]
- Buecker A, Spuentrup E, Grabitz R, et al. Real-time-MR guidance for placement of a self-made fully MR-compatible atrial septal occluder: in vitro test. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 2002; 174:283-285.[Medline]
Top
Abstract
Introduction
