The atrial septal defects (ASDs) occlusion by cardiac catheterization is an alternative to the surgical technique. The use of stretching by balloon is considered as the gold-standard for the ASDs calibration. Three dimensional transesophageral echocardiography (3D-TEE) brings an imaging of the ASD surface and allows the measurement of its maximal diameter. With the goal of a percutaneous ASD occlusion, seventy patients had a simultaneous calibration of the ASDs by 3D-TEE and the balloon method. The mean maximal diameter measured by 3D-TEE was at 20 + 15 mm (range 10-28 mm) compared to the mean balloon-stretched diameter at 22 + 5 mm (range 9-31 mm). The correlation between both methods is good (y = 3.15 + 0.77x, r = 0.8, p < 0.0001). The mean difference between the diameters measured by both methods was 1.85 + 3.17 mm. The ASD visualized by 3D-TEE was single in 54 cases and multiple in 16 cases. The correlation between both methods was excellent in those with a single ASD (y = 1.74 + 0.84x; r = 0.85. p < 0.0001) but it was poor in those with multiple ASD (y = 12.4 + 0.4x; r = 0.45). The calibration by the balloon method which tests the septum resistance is an additive method to 3D-TEE which measures the anatomic diameter of ASD.