Background .– Balloon sizing during percutaneous atrial septal defect (ASD) closure leads to bigger device, extending time procedure and radiation dose but remains the gold standard. Real time-3D-TransEsophagealEchocardiography (RT-3D-TEE) allows diameters and area measures on a 3D view. We assessed the relationships between the occlusive balloon diameter (BD), area and diameters measured using 2D- and 3D-TEE. The effect of ASD shape and the predictive value of the measures in children with ostium secundum ASD were investigated.

Methods .– From 2011 to 2013, we prospectively enrolled 30 children (mean weight 30.9 ± 12.9 kgs max 64 min 18) who underwent transcatheter closure of an isolated ASD under 3D-TEE (3D-matrix array 2–7 MHz TEE probe and iE33 ultrasound machine Philips ^® ). ASD diameters were measured by transthoracic echo (TTE), 2D-TEE and off-line by 3D Multiplanar reconstruction (Qlab software ^® ). ASD area was also estimated by delineating the outlines of the defect on the reconstruction software. The shape of the ASD was assessed visually on the RT-3D-TEE “en face” view and was defined as circular ( n = 16) or ovale ( n = 14). An asymmetry index was calculated by the maximal 3D diameter divided by the minimal 3D diameter (mean 1.4 ± 0.2 min 1 max 1.84). A cut-off of 1.25 was set to distinguish ovale ( n = 8) and circular shape ( n = 22).

Results .– The Amplatzer ^® device number was equal to BD ± 1 mm in 23 cases (76.7%) and higher in the remaining cases. Difference between 3D maximal diameter and BD (2.3 mm ± 4.2 min −9.6 max 12.2, P = 0.0051) was higher in round ASD than in ovale shape (4.8 ± 3.5 vs. 1.4 ± 4.1, P = 0.04). ASD area was well correlated with BD ( r = 0.82, P < 0001). Age, body area, weight, size, and retroaortic rim length were not correlated with the difference between 3D and BD. After multivariate linear regression analysis, ASD area by 3D delineation was the only significant variable for the prediction of BD: BD (mm) = 4.5*ASD area (cm ² ) +11. This formula allow a prediction with < 1 mm difference with the observed BD in ½ and ≤ 2 mm in 2/3 of procedures.

Conclusion .– The relationship between BD and echo parameters are influenced by the ASD shape. ASD area estimated by delineation on a 3D view is the most relevant parameter to estimate the BD. It may be sufficient to guide percutaneous ASD closure without balloon sizing in children.