Author + information
- Fadi J. Sawaya, MD∗ (, )
- Danny H.F. Chow, MD,
- Oscar Millan-Iturbe, MD and
- Ole De Backer, MD, PhD
- ↵∗The Heart Center—Rigshospitalet, Blegdamsvej 9, 2100 Copenhagen, Denmark
In March 2015, the U.S. Food and Drug Administration approved left atrial appendage (LAA) occlusion with the Watchman device (Boston Scientific Corp., Marlborough, Massachusetts) to reduce stroke risk in patients with nonvalvular atrial fibrillation. Outside the United States, the Amplatzer Amulet device (St. Jude Medical, St. Paul, Minnesota) is also available for percutaneous LAA closure. The Amulet device is designed to plug the LAA distally with the lobe and to place the larger disk against the LAA ostium—at the level of the limbus between the LAA and pulmonary vein—to obtain complete sealing (1). Compared with the older generation Amplatzer Cardiac Plug device, the Amulet device has a recessed central end-screw, increased lobe length and sizes, a longer waist, and larger disk diameters (1). Directions for use of the Amulet recommend double antiplatelet therapy ([DAPT]; aspirin and clopidogrel) for 3 months followed by aspirin monotherapy (1).
In a recent article, Sedaghat et al. (2) evaluated Amulet device performance with a follow-up transesophageal echocardiogram in 24 patients (mean follow-up of 11.0 ± 8.2 weeks). Device-related thrombosis (DRT) occurred in 16.7% of patients (n = 4 of 24); 3 of the patients were still on DAPT and 1 was on clopidogrel monotherapy (2). Patients with DRT had a lower left ventricular ejection fraction, larger LA, greater spontaneous echocardiogram contrast, and lower peak LAA emptying velocity as compared to those without DRT. In addition, 3 patients with DRT had a history of LAA thrombus, and all of them had incomplete coverage of the limbus by the Amulet disk. In the accompanying editorial, De Lurgio questions the efficacy of the Amulet device quoting the disturbingly high incidence of DRT and raises concerns on the post-procedural antithrombotic management with only DAPT.
Following these reports, we evaluated the incidence of DRT in 64 successive Amulet implants at our institution. In our practice, all patients implanted with the Amulet device receive DAPT for 3 months followed by aspirin lifelong with a routine cardiac computed tomography follow-up at 3 months to assess for DRT and complete LAA sealing. The mean age was 75 years with a mean CHADSVASc (Congestive Heart Failure, Hypertension, Age 65 to 74, Diabetes Mellitus, Stroke/Transient Ischemic Attack/Thromboembolism, Vascular Disease, Sex) score of 3.4. At 3 months, the DRT rate was 3.1% (n = 2 of 64) in our patient cohort. In those 2 cases with DRT, the Amulet disk did not cover the limbus and thrombi were observed within the untrabeculated part of the LAA ostium between the uncovered portion of the limbus and the Amulet disk.
This Amulet DRT rate of 3.1% is lower than the reported DRT rate of 16.7% by Sedaghat et al. (2). This result is also consistent with a large report (N = 1,047) by Tzikas et al. (3), describing an Amplatzer Cardiac Plug DRT incidence of 4.4% in the 63% of patients that received follow-up transesophageal echo. With the Watchman device, DRT is reported to occur in 4.0% to 5.7% of patients (4)—in addition to differences in device design, patients in the PROTECT AF (Watchman Left Atrial Appendage System for Embolic Protection in Patients With Atrial Fibrillation) trial also received oral anticoagulation for 45 days following a Watchman implantation (5).
In conclusion, complete sealing of the LAA ostium with the Amulet disk should be targeted by means of optimal device sizing and implantation technique, as in both series thrombi were found between the Amulet disk and the uncovered portion of the limbus. These observations strongly suggest that failure of the “pacifier principle” and creation of a cul-de-sac may act as a nidus for thrombi and increase thrombogenicity. Maybe a more individualized post-procedural antithrombotic regimen should be considered, as those patients with a reduced left ventricular ejection fraction, larger LA, high CHADSVASc score, or an Amulet disk placed deeper within the LAA could benefit from a short-term course of an oral anticoagulant.
The Amulet device will be evaluated for safety and efficacy in a noninferiority trial comparing this device with the commercially available Watchman LAA closure device in patients with nonvalvular atrial fibrillation.
Please note: Dr. De Becker is a proctor for St. Jude Medical. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
Amplatzer Amulet LAA Occluder Trial (Amulet IDE); NCT02879448.
- American College of Cardiology Foundation
- Sedaghat A.,
- Schrickel J.,
- Andrié R.,
- Schueler R.,
- Nickenig G.,
- Hammerstingl C.
- Holmes D.R. Jr..,
- Doshi S.K.,
- Kar S.,
- et al.
- Main M.L.,
- Fan D.,
- Reddy V.Y.,
- et al.