To evaluate the accuracy and clinical utility of three-dimensional echocardiography in the assessment of the size and shape of the ventricular septal defect in double inlet left ventricle.

We validated the technique in an autopsy study, and then performed a clinical investigation. Six autopsied hearts were immersed in a waterbath and examined with 3-dimensional echocardiography. We identified the cross-section within the dataset which optimally displayed the ventricular septal defect “en face”, and compared its smallest and largest diameters, as well as its area. The ventricular septal defect was then filled with a silicone sealant and a section prepared for direct measurement. In patients, we measured the diameters and area of the ventricular septal defect in endsystole nad computed the aortic valvar area in endsystole from the cross-section showing the aortic valve “en face”. Ten patients with double inlet left ventricle, aged between 2 and 15 years, were studied using rotational or parallel scanning. All patients had undergone banding of the pulmonary trunk at a mean age of 7 (3–36) days, usually at the time of repair of the coarctation. Two patients had undergone surgical enlargement of the ventricular septal defect prior to echocardiographic examination.

The correlation between the areas of the ventricular septal defect in the specimens measured directly and by 3-dimensional echocardiography was r=0.98, with limits of agreement between −0.1– 0.08 cm2. In the patients, the area of the defect was measured as 3.9±2 cm2, whereas the aortic valvar area was 2.6±0.9 cm2. The ratio between the areas was 1.5 (0.5–2.3). Three patients with areas of the ventricular septal defect smaller than those of the aortic valve had resting Doppler gradients between double inlet left ventricle and the aorta of 16, 20 and 30 mm Hgs, respectively.

3-dimensional echocardiography provides accurate assessment of the area of the ventricular septal defect in double inlet left ventricle, and is helpful in identifying patients with subaortic stenosis caused by restrictive defects.

In patients with tricuspid atresia palliated by construction of a total cavopulmonary connection, both pulmonary and systemic circulations depend on the performance of the dominant left ventricle. When estimating the volume of such ventricles using cross-sectional echocardiography, it is necessary to make assumptions concerning the geometry of the ventricular shape. This is avoided by three-dimensional echocardiography, which provides direct volumetric data. Our purpose was to apply this new method to quantify left ventricular volumes and function in patients with tricuspid atresia after construction of a total cavopulmonary connection.

We studied ten patients (median age: 8 years) with tricuspid atresia who had undergone a total cavopulmonary connection, comparing them with 10 normal children matched for age, sex and size. The three-dimensional echocardiography was performed with electrocardio graphic and respiratory gating. A new transthoracic integrated probe designed for small children was used with a rotational scanning increment of 3°. The 60 slices obtained from the ventricular cavity were stored and formatted in a commercial system (TomTec®). End-diastolic and end-systolic volumes, stroke volume and ejection fraction were calculated after manual tracing of the endocardial surfaces. The volumes were indexed to the body surface area.

As seen in the reconstructions, the dominant left ventricle in tricuspid atresia had a spherical shape, whereas the normal left ventricle is oblong. The left ventricular volumes and function in tricuspid atresia were 54±4 ml/m2 (end-diastolic volume), 28±3 ml/m2 (end-systolic volume), 26±7 ml/m2 (stroke volume) and 48±6% (ejection fraction). These volumes were not different from those obtained in the controls (p = NS). The left ventricular stroke volume and ejection fraction in 10 patients with tricuspid atresia were lower than those calculated for the controls (p < 0.05).

Three-dimensional echocardiography provides a quantitative insight into the pathophysiologic function of the dominant left ventricle in tricuspid atresia after construction of a total cavopulmonary connection.