Unusual Spontaneous Repositioning of an Amplatzer Device Embolized into the Left Atrium Resulting in Completed Closure
of Atria

ABSTRACT: The authors described a rare case of spontaneous repositioning of an embolised Amplatzer occluder into the left atrium resulting in complete occlusion of a hemodynamically significant atrial septal defect, in the fossa ovalis, in a 70-year-old man. Only a slight central residual shunting was present, as was shown by transesophageal Doppler echocardiography performed immediately after the procedure, with no apparent shunt at 30, 60 and 120 days after the intervention. J INVAS CARDIOL 2004;16:290–292 Key words: atrial septal defect, Amplatzer, embolisation, transcatheter closure Transcatheter closure of atrial septal defects (ASD) located in fossa ovalis is nowadays frequently used employing a number of different “closure devices” available on the market.^1–5 The Amplatzer occluder (AGA Medical, Golden Valley, Minn.) is the most often used device at this time thanks its integral construction, rapid release screw mechanism and especially due to the possibility of retrieval and safe repositioning of the connected device.⁶ On the contrary, transcatheter retrieval after release of the Amplatzer device is difficult,⁷ therefore surgical or combined interventional surgical means have been used for the removal of the devices embolised into the left atrium or left ventricle.^8,9 Recently published reports by Wilkinson and Goh¹⁰ and Peuster and all.¹¹ proved that transcatheter retrieval of the embolised Amplatzer occluder into the right or left atrium is also feasible. We describe a rare case of “spontaneous repositioning” of the embolised Amplatzer septal occluder into the left atrium in a 70-year-old man with a hemodynamically significant atrial septal defect located in the fossa ovalis. Case Report. A 70-year-old man with chronic atrial fibrillation was admitted to our institution for resting dyspnoea and atrial tachyfibrillation. After the application of amiodarone and beta-blockers, the tachyfibrillation was slowed down and an adequate ventricular response was achieved. Transesophageal echocardiography (TEE) revealed a 15x17 mm ASD at the fossa ovalis, with an adequate rim of septal tissue surrounding the defect, and pulmonary hypertension (Figure 1A). At cardiac catheterisation, a hemodynamically significant left-to-right shunt across the interatrial septal defect was confirmed, with a pulmonary-to-systemic perfusion ratio of^1,9 1.9:1. Interventional closure of the atrial septal defect was attempted. Sizing (Figure 1B) of the defect was performed using Meditech sizing balloon (Meditech, Watertown, Mass.). Both fluoroscopy and TEE guidance were used throughout the sizing procedure to ensure proper positioning of the sizing balloon catheter. The stretched defect diameter was 15 mm. An Amplatzer device with a size of 19 mm was selected for defect closure. The implantation technique is by now well established and has been described in detail previously.^1–3,5,6 Once the device was fully deployed across the ASD, its position and stability were assessed by fluoroscopy and transesophageal echocardiography (Figure 1C). A gentle pushing and pulling of the delivery cable was performed to ensure a secure and stable device position (Minnesota Wiggle). The delivery system was then unscrewed to release the device. Immediately upon release of the device, fluoroscopy and TEE demonstrated a leftward movement of the right atrial disc with embolisation of the Amplatzer device into the auricle of the left atrium (Figure 1D). Our first attempt was to remove the device with a snare catheter (Goose neck, Microvena, White Bear Lake, Minn.) with a snare loop diameter 35 mm placed into a long sheath and remove it completely from the patient. The snare catheter was introduced next to the embolised device and the screw of the device was caught. Attempts failed to retrieve the embolised device through the atrial septum. During the pooling manoeuvre, the device was lost and embolised again into the left atrium. While checking the position of the device by fluoroscopy and TEE, the device was spontaneously turned round, slowly moved along the septum and covered the ASD. The fluoroscopy showed an “optimal position” of completely turned round (the screw mechanism was in the left atrium) device in previous atrial septal defect, no movement of the device was present. TEE showed only a insignificant central residual shunting. We thought this situation to be very unstable and our goal was to pull the device into the right atrium and remove it during open heart surgery. We again tried to catch the device by the snare catheter but once the device “placed itself” into the defect, to introduce any catheter into the left atrium was not possible. Eventually, we decided to remove the device with biopsy forceps (Cordis Corporation, Miami, Florida) but even after the catching the device with biopsy forseps branches, we failed to pull it in the right atrium due to fixed position of the device in the ASD. Furthermore, the occluder was so tightly catched in the septum that it was not able to push it back in the left atrium. Another control fluoroscopy showed a persisting “optimal deployment” of the Amplatzer occluder and TEE was presenting complete occlusion of the ASD with only trivial central residual shunt. At this point we started to think about leaving the device in its present position and avoid surgical explanation. After discussion with the cardiac surgeon we decided to keep the patient on “conservative” management without open heart surgery but with close monitoring at the coronary care unit with everyday transthoracic ultrasound. To confirm a “stable position” of the Amplatzer device the patient was asked to performed a Valsalva manoeuvre and then was encouraged to cough. The stability of the device was also assessed during patient movement from supine to recumbent position. All these manoeuvres were done under fluoroscopic as well as TEE guidance. No movement of the device was present and only slight central shunting was still apparent (Figure 1 E, F). The procedure was concluded and monitoring of the patient at the coronary care unit with everyday transthoracic echocardiography was performed. If there were any signs of instability in the position of the deployed device, it was agreed that the patient would be immediately operated. However, the remainder of the hospitalisation was uneventful and the patient was discharged 14 days after procedure in good clinical condition. Transthoracic ultrasound performed at 30 and 60 days as well as TEE at 120 days showed the optimal device position with no residual shunt present. A decrease in the right ventricular diameter was apparent at 60 and 120 days after the procedure. Discussion. The major advantage of the Amplatzer device is its possibility of retrieval and safe repositioning of the connected device. However, limited information is available in the case of an embolised device. To date, three reports have been published describing transcatheter retrieving of the embolised Amplatzer occluder. Wilkinson and Goh¹⁰ described a successful retrieval of the embolised device in the right atrium into a long sheath employing a snare catheter. On the contrary, Berger et al.9 published their experience with unsuccessful removal of the device embolised into the outflow tract of the left ventricle using a retrieval forceps and large sheath. Recently, Peuster et al.¹¹ described a transcatheter retrieval and repositioning of the Amplatzer device embolised into the left atrium using pigtail catheter, 0.018 in. guide wire and snare catheter. We described a very rare and unusual case of spontaneous repositioning of an Amplatzer device embolised into the left atrium resulting in completed closure of ASD. The cause of the embolisation is not clear and remains disputable. Both TEE and fluoroscopy were showing an optimal position of the device with the atrial septum between both disks. Furthermore, the Minnesota Wiggle maneuver was done without fault. We speculate that, even with an adequate position during TEE, not the whole part of interatrial septum was really between the disks. Another reason could be the to-and-fro movements of the delivery cable prior to the release of the device. The cause why the device “placed” itself in the ASD and completely it covered is unclear and may only be speculated. Probably, the device was caught by a stream of blood going from the left to the right atrium through the defect and by this stream was placed into the atrial defect. Both TEE and fluoroscopy showed an optimal result of this “spontaneous deployment” but the occluder was completely turned round with the screw mechanism in the left atrium. The TEE was showing that at least two thirds of the interatrial septum was between both disks. The decesion was made to keep the patient for at least 14 days at the coronary care unit with everyday transthoracic echocardiography. We suggested that after this time the occluder will be epitelised and became a “part” of interatrial septum. This suggestion has been confirmed by uneventful hospitalisation as well as by event free 3-month follow-up. Furthermore, transthoracic ultrasound performed at 30 and 60 days as well as TEE at 120 days showed the optimal device position without residual shunt present. One can argue it was to risky to avoid the open heart surgery and leave the turned device in the ASD. But at the end of procedure, the device did not show any sign of unstability and manoeuveres performed earlier proved that the occluder is fixed in the defect. There is no report describing similar case in the literature. Only three articles have been published reporting transcatheter retrieving of the embolised Amplatzer occluder. If the occluder is embolised into the left atrium, the crucial is to prevent its movement into the mitral valve orificium. This could be avoid by insertion of larger pig-tail catheter next to the occluder and above the mitral valve. Then we have a time to think what to do. But usually, combined interventional surgical means are used for the removal of the devices . In agreement with others^9,11 we think, we always should try to pull the device into right-sided heart chambers before open-heart surgery. This is the first report describing the unusually spontaneous repositioning of the Amplatzer occluder embolised into the left atrium and resulting in complete closure of atrial septal defect.