Thrombus Formation during Percutaneous Closure of an Atrial Septal Defect with an Amplatzer Septal Occluder

Abstract

Percutaneous closure of an ostium secundum-type atrial septal defect is typically a safe and effective therapeutic option in the presence of significant shunting or paradoxical embolism. Infrequently, however, periprocedural sequelae occur.

Herein, we report the cases of 2 patients, each of whom underwent transcatheter closure of an atrial septal defect with the use of an Amplatzer Septal Occluder under transesophageal echocardiographic guidance. In both patients, acute thrombi formed periprocedurally, despite preprocedural anticoagulation. In patient 1, the infusion of unfractionated heparin for 24 hours prevented the recurrence of thrombus; in patient 2, the thrombus was isolated under the arm of the occluder, and unfractionated heparin was infused. Both patients were asymptomatic and without detectable thrombus after the procedure and at follow-up. These reports highlight a rare early sequela and the importance of transesophageal echocardiographic monitoring during the percutaneous closure of an atrial septal defect.

Atrial septal defect (ASD) is the most common left-to-right shunt diagnosed in adults. The defect should be closed if there is severe right-to-left shunting or paradoxical embolism.¹ The percutaneous closure of ASDs has become a feasible and safe alternative to conventional surgical closure.² Although excellent results have been reported for transcatheter closure with different devices, concerns have arisen about periprocedural sequelae, including early and late thrombus formation.^3,4 Thrombus formation, although uncommon in percutaneous closure, can result in serious sequelae, such as systemic embolism.

Reports of thromboembolic events related to implanted devices are rare.^3,5,6 Herein, we present 2 reports of periprocedural thrombus formation during ASD closure, despite normal preoperative coagulability values in the patients.

Case Reports

Patient 1

In March 2008, a 33-year-old woman was admitted to our department with a 1-year history of exertional dyspnea. She had no systemic or cardiovascular illness and was taking no medication. Results of physical examination were normal except for fixed S₂ splitting. Laboratory findings, including complete blood count, serum biochemistry, and thyroid function, were within normal limits. An electrocardiogram revealed sinus rhythm, and a chest radiograph showed a mild increase in the vascularity of the pulmonary fields. Transthoracic echocardiography (TTE) revealed a secundum-type ASD with a defect diameter of 16 mm; this was confirmed on transesophageal echocardiography (TEE) in bicaval position. The right ventricular diameter was 30 mm in parasternal long-axis view, and the pulmonary artery pressure as measured on TTE was 45 mmHg. The patient was taken to the catheterization laboratory for hemodynamic evaluation. The Qp/Qs ratio was 2.0, so percutaneous intervention was planned.

The patient was given 100 IU of unfractionated heparin (UFH) and was placed under general anesthesia. An 18-mm-diameter Amplatzer® Septal Occluder (AGA Medical, part of St. Jude Medical, Inc.; Plymouth, Minn) was deployed under TEE guidance, and the ASD was closed. Early thrombus formation was detected at the right atrial edge of the device (Fig. 1A), and atrial fibrillation developed at the same time. Unfractionated heparin was continuously infused (serial activated coagulation time, >200 s), with a target activated partial thromboplastin time (aPTT) of 50 to 70 s. Amiodarone was simultaneously infused. There was no clinical sign of pulmonary or systemic embolism, and sinus rhythm was restored in 8 hours. After 24 hours, both infusions were stopped; TEE revealed no thrombi around the device and no residual shunting (Fig. 1B), and the septal occluder appeared to be in complete contact with the septum in all imaging planes, without any arm fracture. Postprocedurally, test results for hypercoagulability were within normal limits. Hypercoagulable states and recurrent fetal miscarriages were not noted in the patient's detailed medical history.

Fig. 1 Patient 1. Transesophageal echocardiography shows A) thrombus formation on the right atrial edge of the Amplatzer septal occluder, and B) no thrombus on the device after unfractionated heparin was infused.

After 48 hours total, the patient was discharged from the hospital with instructions to take warfarin. At her 1-month follow-up, she was asymptomatic, and TEE showed appropriate positioning of the septal occluder without thrombus or residual shunting. She remained on oral warfarin therapy, with a target international normalized ratio (INR) of 2 to 3. At the 6-month follow-up, TEE results were normal. At the patient's last visit in March 2011, she was asymptomatic, and TEE detected no thrombus around the device.

Patient 2

In May 2009, a 45-year-old man was admitted to the emergency department with 2 hours' difficulty in speaking. His medical history included no systemic or neurologic condition, and he was taking no medication. Physical examination revealed nothing unusual except for the slurred speech. Laboratory evaluations of complete blood count, serum biochemistry, prothrombin time, aPTT, INR, protein C activity, total and free protein S, antithrombin III, anticardiolipin antibodies, homocysteine, and factor V Leiden mutation were within normal limits. An electrocardiogram showed sinus rhythm with incomplete right bundle branch block. A chest radiograph was within normal limits. Cranial magnetic resonance imaging (MRI) and MRI angiography revealed acute ischemic lesions of millimetric size in the left side of the precentral gyrus and the superior temporal and frontal lobes. Transcranial Doppler ultrasonography showed right-to-left flow in a waterspout pattern, which raised suspicions of a patent foramen ovale (PFO) or ASD. An interatrial septal aneurysm was detected with the use of TTE, and TEE revealed a secundum-type ASD of 10-mm diameter in bicaval position. The ASD was suitable for percutaneous closure.

Before intervention, we gave the patient 100 IU of UFH and 300 mg of aspirin and placed him under general anesthesia. When we passed a 13-mm-diameter Amplatzer septal occluder into the left atrium, we observed thrombus formation at the tip of the long delivery sheath (Fig. 2A). Initial attempts to aspirate the thrombus by creating negative pressure with a 50-cc injector were unsuccessful. Because of the risk of releasing systemic emboli, we did not remove the delivery catheter, but instead confined the thrombus in an area between the interatrial septum and the left atrial arm of the occluder before deploying the occluder under TEE guidance. We immediately started anticoagulation by UFH infusion (serial activated coagulation time, >200 s). After the occluder was deployed, TEE revealed no thrombus around it (Fig. 2B). Postprocedurally, the patient was asymptomatic and without neurologic sequelae. He was discharged from the hospital after 48 hours with instructions to take oral anticoagulants for 6 months (target INR, 2 to 3). As of May 2011, he was asymptomatic, and TEE detected no thrombus around the device.

Fig. 2 Patient 2. Transesophageal echocardiography shows A) thrombus formation at the tip of the delivery sheath during percutaneous atrial septal defect closure, and B) no thrombus around the occluder device postprocedurally.

Discussion

In terms of safety and technical simplicity, transcatheter closure of ASDs has become an excellent alternative to surgery, with shorter hospital stays, low morbidity, and very low mortality rates.² Of our 2 patients in whom thrombus formed during percutaneous ASD closure with the Amplatzer septal occluder, one required anticoagulative therapy only. In the other, we isolated the thrombus with an arm of the device and gave subsequent anticoagulation.

In a study of 1,000 patients who underwent percutaneous closure of ASD or PFO,⁷ the reported incidence of device thrombosis was 1.2% in ASD patients and 2.5% in PFO patients. The thrombus was usually observed 4 weeks after the procedures, during scheduled TEE. However, in contrast with other devices, no thrombus formation was observed when the Amplatzer device was used. Postprocedural atrial fibrillation and persistent atrial septal aneurysm were significant predictors of thrombus formation, but device size was not associated with the amount of thrombin activation.⁷ In a review of 54 reported cases of device thrombosis after percutaneous intervention, all types of commercially available closure devices were found to be associated with thrombosis, which was diagnosed at a mean time of 5 months after device deployment (range, immediately–24 mo).⁸ The extensive reports of delayed thrombotic development notwithstanding, thrombus formation can also be observed during intervention, on either the delivery sheath or the device. The Amplatzer system is composed of nitinol and Dacron elements that induce coagulation and platelet-activation systems due to a polyester mesh inside the waist of the device.⁹ In the past, anti-heparin therapy with protamine immediately after intervention was thought to increase thrombogenesis,⁷ but this issue was not clear because of conflicting results from other studies.⁸ Patients who underwent transcatheter closure of ASDs with the Amplatzer septal occluder experienced a significant increase in the coagulation cascade, which peaked at the end of the 1st week without a clear effect on platelet activation.¹⁰ Those findings were confirmed by other investigators,¹¹ who suggested that increased thrombin generation soon after the procedure is most likely related to the deposition of fibrin at the interface between blood and the device. Although the activation of coagulation systems is integral to the sealing properties of the device, those data also raised the question of whether antiplatelet or anticoagulative therapy is optimal after percutaneous closure of an ASD or a PFO.¹¹

During percutaneous ASD closure, thrombus formation on either the delivery sheath or the device is infrequent, and little management guidance is available. Thrombus that has attached itself to the Amplatzer septal occluder has been treated with UFH and subsequent anticoagulation.^6,12 After a PFO closure, a device-related thrombus was treated with recombinant tissue plasminogen activator and a glycoprotein IIb/IIIa inhibitor, and complete thrombus resolution occurred within 48 hours.¹³ Heparin and abciximab were used successfully to treat thrombus on the left atrial disc of an Amplatzer septal occluder.¹⁴ Another method, the suction of thrombus into the delivery sheath from the tip of the transport catheter, requires close monitoring under TEE guidance.¹⁵

Our patient 1 had no risk factors associated with thrombus formation. Patient 2 had interatrial septal aneurysm as a risk factor; however, the thrombus seen at the catheter tip after it passed through the left atrium might have formed inside the delivery sheath. Larger sheaths or shorter procedural times might prevent thrombus formation in such instances. The mobility of acutely formed thrombus during intervention is associated with a high risk of embolization and requires immediate, aggressive therapy. In our patients, we infused UFH and supplemented it with oral anticoagulation, an accepted combined therapy.¹⁵ To minimize the risk of systemic embolism in patient 2, we left the delivery catheter in place and confined the thrombus with the arm of the occluder. Both patients were asymptomatic postprocedurally, and the UFH infusions apparently eliminated the thrombus.

These cases highlight an acute sequela and the importance of close periprocedural monitoring with TEE. Although percutaneous ASD closure is typically safe and effective, early thrombus formation can complicate the intervention and necessitate immediate management.