This article evaluates a new compliant balloon catheter for sizing of Atrial Septal Defects (ASDs) in vitro and in vivo using X-ray and echo measurements. A phantom consisting of a plastic plate with 17 circular holes varying from 8 to 30 mm was used to determine the accuracy of the measurements. The catheter has a 7 Fr shaft with a double lumen. The balloon is made from a 0.002" plastic membrane. Four platinum marker bands on the shaft of the catheter allow radiographic measurements. The balloon catheter was placed through various holes of the phantom and inflated with 38% contrast medium. X-ray measurement by video and cine recording were performed and compared with the true measurement of the model. The study was carried out in a double blind fashion. For echo measurements, the plastic model was immersed in a water bath. The balloon was inflated with water and measured in a long-axis view. The actual diameters were compared with the measurements obtained by video, cine, and echo recordings. One young pig with a surgically created ASD was used for in vivo measurement. The balloon catheter was passed through the ASD over an exchange wire, inflated with diluted contrast medium until a waist was observed, which was measured by video and cine techniques. The animal was euthanized, and the defect was measured by a tapered measuring probe. The difference between the actual size and measurements obtained by video recordings was 0.54 +/- 0.30 mm, by cine 0.62 +/- 0.20 mm, and by echo measurements 0.60 +/- 0.43 mm. All measurements were less than 1 mm in error except for four measurements (<1.5 mm). A strong correlation was present between each two groups (r = 0.99, P < 0.0001). In the in vivo study, the size of the video and cine recordings (14.2 and 13.9 mm, respectively) were very closer to the actual size (14 mm) as measured by a probe in the postmortem specimen. The new balloon catheter provides a more rapid and precise determination of the stretched diameter of atrial septal defects. It eliminates the error of pulling a catheter through the communication. This new technique should be of benefit to all investigators using septal occlusion devices and will facilitate the selection of the properly sized Amplatzer septal occluder.