A persistent PDA does not occur reliably in animals; therefore, a model needed to be created to test the efficacy of the Amplatzer Ductal Occluder. An asynthetic tube was sewn between the descending aorta and the pulmonary artery in animals [22]. The Amplatzer Ductal Occluder device worked well in these animals, completely occluding the surgically created PDA with endothelialization within 3 months. In animal trials, there were no complications, the device was retrievable, and there were no residual hemodynamic abnormalities.

Subsequent to the successful animal studies, clinical trials were undertaken with a high rate of success in occluding various sizes of PDAs in patients [23]. Yet with more extensive clinical use, a few case reports began to surface of partial obstruction of the descending thoracic aorta by the retention disc of the Amplatzer Ductal Occluder device. This was the result of the angle (approximately 65°) of insertion of the naturally occurring PDA with the descending aorta. In these cases, the superior portion of the retention disc was shown to be angled into the aorta by the plug within the PDA (Fig. 37.4), sometimes producing partial aortic obstruction in smaller patients with larger PDAs. It should be noted that the surgically created experimental PDA in the animal model was sewn at a 90° angle to the aorta, so this complication was not predicted by the anatomically inaccurate orientation.

To reduce the complexity of manufacturing the Amplatzer Ductal Occluder device and, at the same time, address the issues of the angle of the descending aorta with the PDA, a new device was designed with an angled retention disc; the wire count was also increased from 72 to 144 wires, thus eliminating the need for fabric in the device [24]. For animal testing of this new design in next stage trials, a synthetic tube was sewn between the pulmonary artery and aorta at the angle of the more naturally occurring human PDA. Subsequently, in all animals in this preclinical trial, the angled Amplatzer Ductal Occluder device was successful in occluding the created PDA (Fig. 37.5A); however, the initial human use of the device revealed shortcomings with these animal trials. Specifically, an angled Amplatzer Ductal Occluder was successfully placed in a young infant with initial success, which also avoided any protrusion of the device into the aorta; however, after 3 months, the device expanded the PDA resulting in “recanalization” (Fig. 37.5B) [25]. This complication was considered to result from two unanticipated design problems. First, the hoop strength of the angled Amplatzer Ductal Occluder was sufficient to dilate the PDA in a young infant. This, in turn, resulted in an increase in the distance between the radial “spokes” of the device, with less occlusive resistance. In retrospect, the animal model trial used a non-expandable PDA with small diameters that limited the distance between the wires of the device.

The latest iteration of the Amplatzer family of ductal occluder devices (Amplatzer Ductal Occluder device 2—ADO2) is longitudinally symmetrical with retention discs on each end and with flexibility at the connection of the retention discs to the occluding plug (Fig. 37.6). This allows the retention discs to angle and lie smoothly against the aortic or pulmonary wall. To create an animal model for testing this device, infant piglets had balloon dilation of the probe PDA so that an anatomically true defect was available [26]. Subsequently, the size of the discs was further reduced to allow implantation in smaller infants (ADO2-Additional Sizes, ADO2-AS). This device was tested after ductal dilation in newborn piglets weighing 1800–2200 g and implanted under echocardiography to mimic occlusion in a premature infant in the isolette, similar to surgical ligation in this population. The animal testing was successful [27], and there are early positive results in clinical trials of premature infants outside the United States [28].