Patient Population

This was a retrospective review of routinely available data sourced from The Prince Charles Hospital echocardiographic database (ProSolv CardioVascular Analyzer; ProSolv CardioVascular, Indianapolis, IN). Information was obtained from a total of 438 individuals; patient demographics are summarized in Table 1 . Group 1 consisted of 219 patients (134 women; mean age, 29.7 ± 25.2 years) with previously diagnosed echocardiographic findings of isolated ASD and left-to-right shunting with no concomitant lesions. Group 2 consisted of 219 age-matched normal controls (119 women; mean age, 29.7 ± 25.2 years) with no cardiovascular pathology and normal echocardiographic findings.

Of the 219 patients (group 1) with secundum ASDs, 101 (46.1%) of these (57 women; mean age, 38 ± 21 years) were deemed suitable and underwent the implantation of percutaneous ASD devices. The remainder of the patients were treated medically (37.9%) because of small lesions or via surgical means (16.0%). RAI measurements were obtained before and after percutaneous ASD closure as well as in the early (mean, 124 days) and late (mean, 390 days) stages of follow-up.

Echocardiographic Data

Patients underwent standard transthoracic echocardiographic examinations according to the American Society of Echocardiography guidelines, using the echocardiographic parameters presented in Table 2 . Studies were performed using commercially available ultrasound systems (Philips iE33, Philips Medical Systems, Andover, MA; Hewlett-Packard 7500, Philips Medical Systems, Andover, MA; Vivid 7, GE-Vingmed Ultrasound AS, Horten, Norway; Acuson Sequoia, Siemens Medical Solutions, Mountain View, CA) and were performed by fully trained and experienced cardiac sonographers.

From the apical 4-chamber, view the RAI was derived by measuring the left atrial (LA) and RA areas. As per the American Society of Echocardiography guidelines, the atrial areas were measured at the end of ventricular systole, when the atrial chambers were at their greatest dimension. When measuring LA area, the pulmonary vein inlets and LA appendage were excluded. The RAI was then derived by dividing the RA area by the LA area ( Figure 1 ).

The RAI is derived by dividing RA area by LA area. (A) Normal example. (B) Example in a patient with a secundum ASD with right-heart dilatation.

RV size was initially measured from two-dimensional guided M-mode tracings performed from the parasternal long-axis view of the left ventricle. The RV cavity was measured at end-diastole from the endocardial surface of the anterior RV wall to the anterior endocardial surface of the interventricular septum. From the apical 4-chamber view, RV size was also assessed qualitatively. The RV was normal in size if the midcavity diameter was smaller than the left ventricle, mildly dilated if the RV diameter was approaching the same size as the left ventricle, moderately dilated if the RV diameter was similar to the left ventricle with sharing of the apex of the heart, and severely dilated if the RV diameter was greater than the left ventricle with the RV forming the apex. RV systolic function was estimated qualitatively on the basis of the displacement of the tricuspid annulus and the function of the RV free wall.

In the absence of RV outflow tract obstruction or pulmonary valve stenosis, RV systolic pressure was estimated via the tricuspid regurgitant jet using the simplified Bernoulli equation: RV systolic pressure = 4( V _TR ) ² + RAP, where V _TR is the velocity of the tricuspid regurgitation signal, and RAP is RA pressure. RAP was estimated via observation of the caliber and reactivity of the inferior vena cava with respiration from the subcostal window. If the inferior vena cava was normal in size (≤1.7 cm), with a 50% decrease in size, RAP was estimated at 0 to 5 mm Hg. Once the inferior vena cava became dilated (>1.7 cm), with a ≥50% inspiratory collapse, RAP was estimated at 6 to 10 mm Hg; with a <50% inspiratory collapse, RAP was estimated at 10 top 15 mm Hg; and with no collapse, RAP was estimated at >15 mm Hg.

Device Implantation

The Amplatzer secundum ASD closure device and delivery system have been described in detail in previous reports. Devices were implanted in the cardiac catheterization laboratory under fluoroscopy and transesophageal echocardiographic (TEE) or intracardiac echocardiographic (Acuson Sequoia) guidance.

Statistical Analysis

Data were collected and statistically analyzed using SPSS version 16 (SPSS, Inc, Chicago, IL). Numerical variables are expressed as mean ± SD. Means were compared using the unpaired t test. A P value ≤ .05 was considered significant. Receiver operating characteristic curve analysis was used to evaluate the ability of the RAI to detect the presence of an ASD. An optimal cutoff value was chosen that provided the best clinical compromise from the derived sensitivity and specificity.