The relation between atrial septal defect shape, diameter, and area using three-dimensional transoesophageal...

A trans-catheter closure of an atrial septal defect (ASD) is efficient. Balloon sizing (BS) during the catheterization leads to an overestimation of ASD size. Three-dimensional transoesophageal echocardiography (3D-TEE) allows the ASD morphology to be assessed comprehensively. The aim of this study was to assess the relationships between the shape and the measurements of ASDs by 2D-, 3D-TEE, and BS in children.

Thirty children who underwent percutaneous closures of a single ASD were enrolled. ASD diameters were measured by 2D-transthoracic echocardiography (TTE), 2D-TEE, 3D-TEE and compared with BS. The ASD area was measured on 3D-TEE images after multi-planar reconstruction. ASD was estimated as round or oval on 3D-TEE 'en-face' view. 2D-TTE, 2D-TEE, and 3D-TEE(max) ASD diameters were well correlated with BS (r = 0.75; 0.80, and 0.85, respectively). Mean diameters were all significantly smaller than the mean BS. The mean difference between the balloon area and 3D-TEE area was 1.6 ± 1.4 cm(2) (P < 0.0001). The mean difference between BS and 3D-TEE(max) diameters was higher in round ASDs than in oval ASDs (4.0 ± 3.3 vs. 1.1 ± 3.3, P = 0.02). With multivariate linear regression analysis, two formulas were built to predict BS. The first model was BS = 1.07 × 3D-TEE(max)- 3.1 × ASDshape + 3. The ASD shape was 0 for round and 1 for oval ASDs. A second model was BS = 4.5 × ASDarea + 11.5.

The ASD shape is accurately estimated by 3D-TEE and influences the relationship between echocardiographic measurements and BS. The ASD shape, its maximal diameter and the area assessed by 3D-TEE may be sufficient to determine the device size without BS in children.