By Jeffrey Zimmet, MD, PhD

Associate Professor of Medicine, University of California, San Francisco; Director, Cardiac Catheterization Laboratory, San Francisco VA Medical Center

Dr. Zimmet reports no financial relationships relevant to this field of study.

SYNOPSIS: A large observational study from France of two devices used commonly for occlusion of the left atrial appendage in patients with atrial fibrillation showed that device-related thrombus was not uncommon and was associated with subsequent stroke.

SOURCE: Fauchier L, Cinaud A, Brigadeau F, et al. Device-related thrombosis after percutaneous left atrial appendage occlusion for atrial fibrillation. J Am Coll Cardiol 2018;71:1528-1536.

The use of oral anticoagulation to prevent stroke and other thromboembolic events is an important principle in the treatment of patients with atrial fibrillation (AF). Based on the concept that most left-sided thrombus in AF develops in the left atrial appendage (LAA), occlusion of the LAA by mechanical devices has emerged as an alternative to long-term anticoagulation. In Europe, most of these procedures are performed with either the Watchman nitinol cage device, which has been studied in two large randomized, clinical trials and is approved for use in the United States, or with a version of the Amplatzer Amulet or Cardiac Plug devices, which has data from large European registries. Clinical trials have primarily enrolled AF patients who are fully eligible for anticoagulation to allow for randomization. Trials of the Watchman device have used a post-implant regimen of 45 days of aspirin and warfarin, followed by 4.5 months of dual antiplatelet therapy (DAPT). Most patients in these trials have undergone post-implant imaging by transesophageal echo to look for both device leak and for device-related thrombus (DRT). Clinical trials of these devices have consistently reported low DRT rates — 20 of 478 implants in the combined PROTECT-AF trial and the CAP registry, for example. Additionally, the consequences of DRT have been downplayed both in the literature and in the device packaging, with very low reported rates of embolic events associated with DRTs.

In real-world use, LAA occlusion often is applied to patients who have contraindications to long-term anticoagulation and who are at higher risk of thromboembolism than patients studied in the approval trials. Regimens of post-implant anticoagulation vary, as do follow-up imaging. Fauchier et al examined data from eight centers in France to evaluate clinical outcomes in patients using the two main LAA closure systems, differences in antithrombotic management at discharge from hospital, and the incidence and prognosis of thrombus formation on the device after LAA occlusion.

In their cohort, 469 consecutive patients underwent LAA closure with either the Watchman (n = 272) or ACP-Amulet (n = 197) devices. Mean follow-up was 13 months, and post-implant imaging with either transesophageal echo or CT scans was available in 339 patients an average of 2.8 months post-procedure. DRT was found in 26 of these patients, giving an incidence of 7.7%. While older age and history of stroke were found to be independent predictors of DRT, use of oral anticoagulation or DAPT at discharge were protective. Importantly, DRT was found to convey a significantly higher risk of stroke compared with patients without associated thrombus, with a hazard ratio > 4. The authors concluded that thrombus formation on LAA occluder devices is not uncommon, and that this finding is strongly associated with a higher risk of ischemic stroke during intermediate-term follow-up.


The main findings of this study were that thrombus development on LAA occluder devices in real-world use is higher than what has previously been reported, and that this is associated with increased rates of thromboembolic complications.

Although some would note that this study was limited by its retrospective nature and incomplete imaging, its relatively large size and independent nature (no industry support) make its findings highly relevant to clinical practice. Industry has promoted the concept that DRT is a relatively benign finding, which defies logic. As one would expect from basic principles, this study suggests that thrombus on a device in the LAA confers an increased risk of thromboembolism.

Real-world LAA occluder patients generally are at higher thromboembolic risk compared with those studied in the randomized trials, and often include patients without options for systemic anticoagulation. These patients are problematic, as those who are not treated with oral anticoagulants or with DAPT post-implant clearly are at higher risk for DRT, with subsequent increased risk for stroke. Remember that the Watchman device is considered to be contraindicated when the use of warfarin, aspirin, or clopidogrel are not possible. As noted in the accompanying editorial, a DRT does not represent a failure of LAA occlusion as a preventive strategy, but is instead an iatrogenic complication.

Real-world patients also clearly undergo lower rates of post-implant imaging than what is recommended, whether because they are too frail, or because they refuse post-implant imaging. Follow-up imaging, most often by TEE, is an important part of care after LAA occluder implantation. Anticoagulation should be prescribed in response to a diagnosis of DRT.

Those caring for patients with atrial fibrillation should remember that LAA occlusion is not a complete substitute for anticoagulation, and short-circuiting the recommended post-implant regimen of anticoagulation results in higher rates of DRT. LAA occlusion remains a viable option for many, but patients and their medical providers should be aware of the associated issues.