No CrossRef data available.
Published online by Cambridge University Press: 21 January 2005
We occluded a large arterial duct in a patient with atrial fibrillation, using an Amplatzer duct occluder, with no residual shunt at the end of the procedure. Follow-up revealed recanalization, which was successfully treated with a 4 mm Amplatzer ASD occluder. It is speculated the reopening of the initially occluded duct was the consequence of anticoagulation needed because of the atrial fibrillation.
A patent arterial duct is usually diagnosed and treated in early childhood. In most patients, the shunt across the duct into the pulmonary arteries is small. The reason for closure, in most patients, therefore, is prevention of infective endocarditis.1 Transcatheter closure of the duct can be achieved with several different approaches. The implantation of a single or multiple coils of various sizes is a safe procedure that usually leads to complete occlusion of the persistently patent duct.2, 3 The insertion of coils, nonetheless, is technically demanding, and has limited success for complete closure when the duct is large. The recent introduction of the Amplatzer duct occluder offers the possibility of complete closure by catheter intervention, even for those with very large ducts.4 Patency of the duct in an elderly patient with hemodynamically significant shunting is rare. Here, we report recanalization of a duct initially occluded successfully with an Amplatzer device.
A 71-year-old patient presented with syncope and atrial fibrillation. On physical examination, the patient was obese, without signs of heart failure. The electrocardiogram documented atrial fibrillation with a ventricular rate of 60 beats/min. Echocardiography revealed a moderately depressed left ventricular function, and a dilated left atrium. Transoesophageal echocardiography showed a shunt between the aorta and the pulmonary arteries. This was further analyzed using magnetic resonance angiography, which confirmed the patency of the large duct. At catheterization, we found a substantial shunt across the duct, with a ratio of pulmonary to systemic flow of 1:1.5, with an increased mean pulmonary arterial pressure at 50 mmHg. The duct was occluded using an Amplatzer 12/10-mm duct occluder. Immediately after the procedure, the shunt was no longer detectable by oximetry, with the findings confirmed by transoesophageal echocardiography prior to discharge from hospital.
Subsequent to discharge, the patient was asymptomatic, with an improved exercise tolerance and spontaneous reversal to normal sinus rhythm, for about 1 month. At that time, atrial fibrillation recurred. Transoesophageal echocardiography then showed a new shunt across the duct immediately adjacent to the Amplatzer device, which had not changed its position compared to the preceding investigation. Invasive re-evaluation revealed a shunt of 20% from left to right. This minor shunt was closed with a 4 mm atrial septal defect occluder implanted via an aortic approach (Fig. 1). After the successful insertion of the occluder next to the Amplatzer device, the shunt was no longer present. Follow-up transoesophageal echocardiography, nonetheless, revealed a trivial shunt of 1 mm located inferior to the second device. Throughout the period of observation, the patient was continuously anticoagulated with warfarin, achieving therapeutic targets, in order to prevent possible embolization from atrial thrombus formed during atrial fibrillation.
Figure 1. Fluoroscopic visualization of two devices occluding the arterial duct. An Amplatzer atrial septal defect occluder of 4 mm size (arrow) was introduced 13 months later, after a residual shunt had been detected immediately adjacent to a 12/10-mm Amplatzer duct occluder.
The diagnosis of a patent arterial duct in this septuagenarian was serendipitous, as a large hemodynamically relevant duct is normally unheard of in patients of this age. In our patient, the significant shunt from left to right may have contributed to fibrillation due to atrial distension, since sinus rhythm was restored after the initial successful closure, albeit only temporarily. Because of the beneficial effect, we decided to abolish the residual shunt when fibrillation recurred, and a relapse was detected in terms of shunting. The second occlusion, using an occluder designed for the atrial septum, was not totally successful, as a small residual shunt was present at the inferior edge of the device during follow-up. We deemed this not to be hemodynamically significant.
To the best of our knowledge, the reopening of a duct previously successfully occluded using an Amplatzer device has not been documented.5 We consider a primary technical failure of the initial device an unlikely explanation, since angiographic, echocardiographic, and oxymetric tests all indicated complete primary closure. The reopening of the previously occluded duct suggests anatomic remodeling. It is not clear to what extent this might have been facilitated by the reduced thrombogenesis due to the chronic anticoagulation prescribed for atrial fibrillation.

Fluoroscopic visualization of two devices occluding the arterial duct. An Amplatzer atrial septal defect occluder of 4 mm size (arrow) was introduced 13 months later, after a residual shunt had been detected immediately adjacent to a 12/10-mm Amplatzer duct occluder.