Author + information
- Received February 13, 2001
- Revision received July 10, 2001
- Accepted July 26, 2001
- Published online November 1, 2001.
- ↵*Reprint requests and correspondence: Dr. Mark C. Johnson, Division of Cardiology, St. Louis Children’s Hospital, One Children’s Place, St. Louis, Missouri 63110 USA.
The purpose of this study was to prospectively characterize the reduction in right atrial (RA) area and right ventricular (RV) volume after transcatheter closure of atrial septal defect (ASD) and to investigate factors that may predict magnitude of resolution in right heart enlargement.
Secundum ASD can cause volume overload of the right side of the heart with the potential for development of late complications. Little is known about reduction in right heart size after closure of ASD.
Transthoracic echocardiography was performed in 38 patients undergoing transcatheter closure of ASD. The RA area and RV volume were measured prior (n = 38), within 24 hours (n = 37), at 3 to 6 months (n = 24), at 12 months (n = 20) and at 24 months (n = 10) after closure of ASD. Change over time within the study group was assessed and the study group was compared to a control group of 19 patients with structurally normal hearts.
Indexed RA area decreased from baseline to 3- to 6-month follow-up (p = 0.004) as did indexed RV volume (p < 0.0001). Indexed RV volume was similar to that in the control group at 24 months (p = 0.3); however, indexed RA area remained greater than in the control group (p = 0.006). Decrease in indexed RA area over the first 12 months of follow-up was related to young age at time of closure by regression analysis (r = 0.55, p = 0.013).
Closure of secundum ASD results in decreased indexed RV volume comparable to that in control subjects at 24 months following closure. Indexed RA area remains increased compared to that in control subjects but does decrease over time. Decrease in RA area is inversely proportional to age at time of ASD closure. Long-term follow-up is required to evaluate the clinical impact of persistently increased RA size.
Secundum atrial septal defect (ASD) can cause volume overload of the right side of the heart with the potential for subsequent development of right heart failure, systemic embolism, elevated pulmonary vascular resistance or atrial arrhythmias (1–4). Closure of these defects is recommended to lessen the risk of late complications. The traditional indication for surgical repair of ASD has been the ratio of pulmonary to systemic blood flow as measured at cardiac catheterization. Magnetic resonance imaging (MRI) is the traditional noninvasive method of measuring right ventricular (RV) volume in evaluating the patient with left-to-right shunt. Evidence of right heart enlargement by echocardiographic imaging has gained acceptance as an indication for defect closure as it has been shown to accurately estimate the degree of shunt (5,6). Ellipsoid shell model-based calculations of RV volume using measurements of RV dimensions obtained by MRI have been shown to correlate closely with both dual-plane cine MRI and biplane cineangiography determinations of RV volume (7). Denslow and Wiles (8)have previously reported that echocardiographic measurements applied to this same model correlate closely with MRI-determined RV volume. The purpose of this study was first to characterize the time course of echocardiographic resolution in right atrium (RA) area and RV volume overload in a wide age range of patients undergoing catheterization for device closure of ASD. Second, we investigated factors that may predict the magnitude of resolution in right heart size after ASD closure.
Over a 3-year period from June 1997 through September 2000, all patients enrolled for closure of a secundum ASD or patent foramen ovale (PFO) with the Amplatzer Septal Occluder device (AGA Medical Corporation; Golden Valley, Minnesota) under an investigational protocol approved by the Human Studies Committee at Washington University School of Medicine were eligible for this study. Patients with additional hemodynamically significant cardiac defects were excluded from this study. Echocardiographic controls were obtained from patients with normal two-dimensional, Doppler and color flow studies as a part of a routine clinical evaluation of heart murmurs or chest pain.
Subjects from the device protocol had echocardiograms performed within 24 h preceding cardiac catheterization by one of six different echocardiographic technicians then per the investigational protocol within 24 h after device closure, within 3 to 6 months after device closure, within 12 months after device closure and then at 12-month intervals. An average of three to five determinations of each two-dimensional echocardiographic dimension was measured off-line (echoPro+; Houston, Texas) by one of the authors for each of the echocardiograms obtained. The RA area and ellipsoid shell RV volume were determined according to methods described in earlier studies (5,8). The RA area and RV volume were indexed to body surface area.
Cardiac catheterization was performed with the patient under general anesthesia and inspired oxygen concentration of <25%. Pulmonary and systemic flows were determined by the Fick method. Right and left heart catheterization was performed and pressures were determined with fluid-filled catheters. Amplatzer Septal Occluder devices were placed under a specific protocol. Transesophageal echocardiography was used as part of this protocol to confirm stretched diameter of the secundum ASD as well as to verify placement of the device.
Statistical analysis was performed with StatView for Windows, version 5.0.1 and SAS version 8.1 (SAS Institute, Inc; Cary, North Carolina). Change over time in the study group was assessed by repeated measures analysis of variance (ANOVA) up to 12-month follow-up. Regression analysis was performed studying change in indexed RA area and indexed RV volume over the first 12 months of follow-up in relationship to rank age at time of device placement and baseline pulmonary blood flow: systemic blood flow (Qp:Qs). Control and baseline study group demographics were compared with two unpaired ttests. Study group indexed RA area and RV volume were compared to RA area and RV volume in control subjects at baseline and 24 months with analysis of covariance (ANCOVA), adjusting for age at time of closure.
Over the 3-year study period, 59 eligible patients underwent catheterization for possible device closure of secundum ASD or PFO. Of these patients, five had fenestrated ASDs and did not undergo device closure. Patients who received the device but who were eliminated from the study group were those with other, hemodynamically significant, cardiac defects (n = 9) and those who had baseline Qp:Qs<1.5 (n = 7). Thirty-eight patients were enrolled in the study group. The control group included 19 patients who were evaluated for murmur or chest pain and had structurally normal hearts with an assumed Qp:Qs of 1.
Baseline demographic and clinical data for control and study groups are shown in Table 1. There was no statistically significant difference between the control and study groups in gender (p = 0.2), age (p = 0.2) or weight (p = 0.5). Small residual shunts (<2 mm) were identified in 12 patients (32%) at 24 h and in 4 patients (11%) at 1 year. Measurements of indexed RA area and indexed RV volume obtained from the study group over time are shown in Table 2. Analysis by repeated measures ANOVA demonstrated that study group indexed RA area decreased significantly from baseline to 3 to 6 months (p = 0.004). The change in RA area by this analysis was strongly related to age at time of closure (p = 0.0013) and time of follow-up (p < 0.0001) with an interaction of age and time (p = 0.0023). The change in RA area was greatest and persisted farther into the follow-up period in the youngest patients. Z scores illustrating the change in study group indexed RA area over time are shown in Figure 1. Indexed RV volume also decreased significantly from baseline to three to six months (p < 0.0001). There was no interaction of age and time in this analysis (p = 0.17). Z scores illustrating the change in study group indexed RV volume over time are shown in Figure 2. Indexed RA area did not change significantly from 3 to 6 months to 12 months (p = 0.43), nor did indexed RV volume (p = 0.054).
Regression analysis showed a relationship between rank age at the time of ASD closure and change in indexed RA area as seen in Figure 3. Younger age at time of device closure was associated with greater change in indexed RA area (r = 0.55, p = 0.013). There was no relationship between baseline Qp:Qs and change in indexed RA area (r = 0.14). There was no relationship between change in indexed RV volume and age at closure or baseline Qp:Qs (r = 0.17, r = 0.19).
The ANCOVA revealed that the study group had greater indexed RA area, adjusted for age at time of closure, compared to control values at baseline (p < 0.0001) and after 24 months (p = 0.006). The study group also had greater indexed RV volume compared to control values at baseline (p < 0.0001). Indexed RV volume was similar to control values at 24 months (p = 0.3).
While long-term functional status of children who have undergone surgical closure of hemodynamically significant ASD has been shown to be good (9), there is no demonstrated benefit from early versus late closure based on a study of exercise performance (10). There may be, however, a risk for other complications such as right heart failure, systemic embolism, elevated pulmonary vascular resistance or the development of atrial arrhythmias, when closure of secundum ASD is delayed.
Echo demonstration of RA area and RV volume resolution
This study demonstrates a reduction in right heart size as early as three to six months following device closure of secundum ASD, and likely even earlier than that. Despite this significant decrease in right heart size, indexed RA area remains increased compared to that in control subjects 24 months after closure. Our study also demonstrated that the degree of reduction in indexed RA area was related to young age at time of device placement. Presumably, the early reduction in right heart size is secondary to removal of the left-to-right shunt and reduction of preload. Further changes over the next 18 months likely occur as remodeling of the myocardium takes place. Assessment of myocardial mass and ejection fraction may help elucidate the mechanisms of these delayed changes as well as the impact of young age at ASD closure on resolution of right heart enlargement.
Recent data presented as an abstract showed a decrease in a single echocardiographic dimension, the right ventricular end-diastolic dimension (RVEDD), as soon as 24 h after device closure of ASD in children (11). However, the RVEDD does not correlate well with RV volume because of the complex geometry of the RV. For this reason, we used the ellipsoid shell model that has been shown to correlate well with MRI measurements of RV volume.
RA dilation and late complications
Right heart size in adults with ASD was demonstrated by cardiac catheterization to decrease in a series of 12 patients following surgical closure (12). Studies of long-term outcomes in adults demonstrate contradictory findings when functional class and survival are examined (12–15). However, the incidence of new atrial arrhythmias in adults is consistently unchanged following surgical closure compared to those treated conservatively (12,14,15), and some studies suggested that older age is a risk factor for persistent atrial arrhythmias and development of new atrial arrhythmias following surgical closure of ASD (16,17). The etiology of late atrial arrhythmias following surgical closure of ASD in adults is not well explained. Long-standing volume overload, varying degrees of pulmonary hypertension, ventricular dysfunction, and congenital defects in the atrial conduction tissue have all been implicated (18,19). Animal and human models have demonstrated that stretch or dilation causes changes in the electrophysiologic characteristics of atria making them vulnerable to arrhythmia (20,21). In addition, treatment of atrial arrhythmias in the patient with dilated atria is less successful than in the patient with normal atria (22–24). Studies in children with secundum ASD have identified prolonged atrial conduction and sinus/atrioventricular node dysfunction (25,26)and one study showed improvement in atrial conduction disturbances after transcatheter closure of ASD (19). The present study suggests that change in the RA area was related to age, but that on the whole, RA area does not return to normal over 24 months following ASD closure by transcatheter technique. It is possible that the atria of the youngest patients have greater remodeling potential and that this patient population may be at less risk for the development of late atrial arrhythmias.
The present study has some limitations. First, the number of patients we have followed to 24 months may be too small to demonstrate persistently increased RV volume. Second, device placement across the secundum ASD may alter the geometry of the RA and prevent its return to a normal dimension. Finally, the precise effect of small residual shunts in four study group patients at 12-month follow-up is not defined. We assumed that there would be no significant hemodynamic consequence of an atrial level shunt <2 mm.
Further study is warranted to quantify the risk of late arrhythmias following transcatheter closure of secundum ASD in order to determine an optimum age for closure. Additional patients should be studied to verify the return of RV volume to baseline following closure and also to determine the geometric changes that occur in the RA with the placement of the ASD device. Finally, the relative risks for development of complications following device closure of ASD versus surgical closure need to be explored.
We thank K. M. Trinkaus, statistician in the Division of Biostatistics, Washington University School of Medicine, for her assistance in the statistical analysis of this study.
↵1 IRB approval and informed consent were both obtained for placement of the Amplatzer Septal Occluder device.
- analysis of covariance
- analysis of variance
- atrial septal defect
- magnetic resonance imaging
- patent foramen ovale
- pulmonary blood flow
- systemic blood flow
- right atrium
- right ventricle
- right ventricular end-diastolic dimension
- Received February 13, 2001.
- Revision received July 10, 2001.
- Accepted July 26, 2001.
- American College of Cardiology
- Campbell M
- Steele P.M,
- Fuster V,
- Cohen M,
- Ritter D.G,
- McGoon D.C
- Haworth S.G
- Rosenthal M,
- Redington A,
- Bush A
- ↵O’Brien SE, Zhu W, Cao QL, Hijazi ZM. Rapid decrease of right ventricular volume overload following transcatheter closure of secundum atrial septal defects (abstr). J Am Coll Cardiol 2000;35 suppl A:510.
- Favilli S,
- Zuppiroli A,
- Mori F,
- Santoro G,
- Manetti A,
- Dolara A
- Shah D,
- Azhar M,
- Oakley C.M,
- Cleland J.G.F,
- Nihoyannopoulos P