|Home | About | Journals | Submit | Contact Us | Français|
During transcatheter occlusion of a patent ductus arteriosus, one potential complication is that the coil can embolize into one of the branch pulmonary arteries or the aorta. It is often possible to remove this coil percutaneously, but at times, surgical intervention is required. The present report describes a case in which the coil migrated to the left pulmonary artery and repeated attempts to retrieve the coil were unsuccessful. A left thoracotomy was performed, the coil was removed and the patent ductus arteriosus was ligated.
Transcatheter closure of a patent ductus arteriosus (PDA) using a coil has been performed since 1967 and is considered to be a safe alternative to surgical ligation (1). Since the first coil closure, the technique has been improved such that the majority of PDAs can be safely closed (1). One of the most common complications of transcatheter closure of the PDA is embolization to the branch pulmonary arteries or the aorta (2). In the majority of cases, the embolized device can then be retrieved percutaneously; however, we present a case in which the coil embolized to the left pulmonary artery and surgical removal was required.
A 16-month-old girl was found to have a continuous murmur at the upper left sternal border and was diagnosed by echocardiography as having a 3.5 mm PDA with left-to-right shunting. It was decided that she would undergo coil occlusion of the ductus arteriosus because her anatomy was found to be favourable for transcatheter closure (Figure 1). Her right femoral vein and right femoral artery were cannulated, and a 5 Fr catheter (Cook Canada Inc) was advanced into the right femoral vein. A 6.5 mm × 5 mm loop coil (Cook Canada Inc) was delivered but it did not sit properly in the aortic ampulla and, therefore, was removed. A 5 mm × 5 mm loop coil (Cook Canada Inc) sat comfortably in the ductus arteriosus. There was residual flow around the device at first; however, this decreased after 5 min. A stability test confirmed that the coil was in a good position; however, just as the wire was being rotated to release the coil, the coil rotated and migrated to the left pulmonary artery. The coil lay parallel to the vessel and despite multiple attempts at using the snare catheter, it could not be retrieved. A sheath was placed and a bioptome was manoeuvred adjacent to the coil, but the device still could not be removed. At this point, the distal left pulmonary artery was found to be essentially occluded and, therefore, surgical removal of the coil was deemed necessary (Figure 2). A left thoracotomy was performed and the left pulmonary artery was dissected out within the major fissure. Just distal to the origin of the lingular branches, the PDA coil was identified within the artery, and the proximal vessel was snared with a vessel loop and the distal vessel clamped. A transverse arteriotomy was made, the coil was removed, and proximal and distal back bleeding confirmed. The vessel was repaired with a single layer of 7-0 prolene sutures and the ductus arteriosus was doubly ligated with a heavy silk tie. The chest was then closed in the usual fashion. The patient recovered well from surgery and was discharged from the hospital on day 3. At follow-up six months later, no evidence of a residual ductus arteriosus or of branch pulmonary artery stenosis was identified.
Possible complications of catheter-based PDA closure are coil migration, coarctation, left pulmonary artery stenosis and, very rarely, hemolysis (3). Residual shunts immediately after placement of the coil are also common (4), although the majority later spontaneously thrombose (3,5). Embolization of the coil can occur regularly, with frequencies of up to 16% being reported (3). Removal by snare catheter is preferred and can often be accomplished (2,4). In cases in which this is not possible, the coil may be left in place if it is unlikely to cause further problems. When the coil location results in occlusion of a major vessel, surgery becomes necessary (2,4,5).
The coil usually migrates to the pulmonary artery because of incorrect deployment or underestimation of the size of the ductus (5). In total, we have seen three coil embolizations, of which two were successfully retrieved and subsequent coils were placed in the proper position. The third case is the subject of the present case report whereby the placement was tested for stability, which implied that the coil was secure. After it had moved, the position of the coil, which lay parallel to the wall of the left pulmonary artery, made removal by catheter impossible. Because perfusion to the distal left pulmonary artery was significantly decreased, it was not possible to leave the coil where it was, and surgical removal was required.
A left thoracotomy was performed in this patient to remove the coil and ligate the PDA. This was advantageous because it allowed us to avoid the potential complications of cardiopulmonary bypass and gave us an excellent view of the ductus arteriosus and the coil, compared with a median sternotomy.
At our centre, contraindications for coil occlusion include patients weighing less than 5 kg with a ductus arteriosus larger than 4 mm to 5 mm in diameter; small patients with a ductus arteriosus larger than 4 mm to 5 mm and a shallow ampulla, in which the Amplatzer device (AGA Medical Corp, USA) aortic disc would protrude into the descending aorta; a long, tortuous ductus arteriosus; and a type D (tubular) ductus arteriosus 5 mm or larger in a small patient. In the last case, a larger Amplatzer device is likely required and there is a higher tendency for the torque from the cable to pull the device through the ductus arteriosus. If the device is placed in a higher position, the likelihood of the aortic disc projecting into the descending aorta increases.