Transcatheter left atrial appendage closure for stroke prevention among atrial fibrillation patients

Keywords::

• atrial fibrillation
• left atrial appendage
• left atrial appendage occlusion

Stroke is the most feared complication of atrial fibrillation (AF) and remains a major cause of serious disability and death in these patients. Therefore, strategies to prevent stroke among AF patients are essential and can effectively improve prognosis [1,2]. The standard stroke prevention among AF patients consists of life-long oral anticoagulation mainly using vitamin K antagonists. Having, at a maximum, a 67% relative risk reduction compared with placebo [3], in addition to various intrinsic limitations and adverse effects (e.g., narrow therapeutic range, variability in pharmacokinetics, food dependency of efficacy or 0.5% annual risk of cerebral hemorrhage), vitamin K antagonists represented for many decades a suboptimal ‘best medical therapy’ for AF patients. Novel oral anticoagulants (e.g., factor Xa antagonists and direct thrombin inhibitors) with better safety, efficacy and compliance than warfarin are therefore much welcomed [4,5]. Yet the long-term outcomes of these products are unknown, and bleeding, although reduced, will not be fully eliminated. This excludes certain patients even from novel oral anticoagulants.

Transesophageal echocardiographic, surgical and autoptic studies have shown that the left atrial appendage (LAA) is the site for thrombus formation in 90% of patients with thrombosis owing to nonvalvular AF [6–8]. This finding formed the rationale for LAA closure procedures as a stroke prevention therapy among AF patients. This was initially done by excision or ligation as part of open-heart surgeries [9], later thoracoscopically [10], and since 2001 via a transcatheter approach [11].

History of transcatheter LAA occlusion procedures

The percutaneous LAA occluder (PLAATO, EV3 Endovascular, Inc., MN, USA) was the prototype device for this procedure, and after demonstrating safety and feasibility in animal studies (revealing complete LAA occlusion, no evidence for thrombi on the implant surface and complete healing 3 months after device implantation), the world’s first percutaneous LAA occlusion in man was performed on the 30 August 2001 by Horst Sievert and Michael Lesh in Frankfurt, Germany [11]. A few hundred PLAATO devices were implanted in the setting of small clinical series and one multicenter, prospective observational study [12]. In 2006, the company withdrew the device short of the financial means projected for clinical approval. On 10 April 2002, the world’s first percutaneous LAA occlusion in an awake patient (local anesthesia only without transesophageal echocardiography guidance) was performed in Bern (Switzerland). A multicenter registry investigated the Amplatzer technique for percutaneous obliteration of the LAA by implanting Amplatzer devices, originally used for closure of a patent foramen ovale (PFO) or an atrial septal defect (ASD), in the LAA, thereby taking advantage of the ease of use and low thrombogenicity of the these devices [13]. Continuous refinement of the Amplatzer technique led to the development of the dedicated Amplatzer LAA occlusion device system, the Amplatzer Cardiac Plug (ACP; St. Jude, MS, USA), which received CE mark approval in December 2008. The WATCHMAN device (Boston Scientific Atritech, Inc., MN, USA), the third device line introduced, was first implanted on 22 August, 2002, in Leipzig, Germany [14]. In 2009, the first prospective randomized trial (PROTECT AF: WATCHMAN Left Atrial Appendage System for Embolic Protection in Patients with Atrial Fibrillation [15]) for transcatheter LAA closure compared with warfarin was presented and proved noninferiority of transcatheter LAA closure to standard oral anticoagulation with an emerging advantage with growing experience and follow-up duration. This paved the way for the clinical adoption of the technique. ACP and the WATCHMAN device will have to complete their ongoing US pivotal (ACP) or complementary (WATCHMAN) trials before they can achieve full US FDA approval. Currently, a number of additional transcatheter devices are at different stages of clinical evaluation, for example SIDERIS transcatheter patch (Custom Medical Devices, Athens, Greece) [16], WaveCrest device (Coherex Medical, UT, USA) and Figulla device (Occlutech, Helsingborg, Sweden).

Effectiveness of transcatheter LAA device closure for stroke prevention

The highest level of evidence was delivered by the sole randomized trial in the field so far. The multicenter prospective PROTECT AF trial randomized 707 patients in a 2:1 manner nonvalvular AF, and one or more risk features for systemic embolism (previous stroke, transient ischemic attack, congestive heart failure, diabetes, hypertension or age ≥75 years) to percutaneous closure of the LAA or chronic warfarin therapy with a target INR of 2–3. The study investigated noninferiority of the WATCHMAN device regarding effectiveness and safety. The primary end point for effectiveness was a composite of stroke, cardiovascular death and systemic embolism. At 1065 patient-years of follow-up, the primary efficacy event rate was 3.0 per 100 patient-years (95% credible interval [CrI]: 1.9–4.5) in the intervention group and 4.9 per 100 patient-years (CrI: 2.8–7.1) in the control group (rate ratio: 0.62; 95% CrI: 0.35–1.25). The probability of noninferiority of the intervention was >99.9% [15].

Safety of transcatheter LAA device closure interventions

Safety concerns are either procedure related, which is basically the same for all devices and is commenced by a femoral venous puncture followed by transseptal access to the left atrium (via a needle puncture or an existing PFO or ASD), or device related where differences exist between delivery sets and implants. In PROTECT AF, the primary end point for safety included major bleeding, pericardial effusion and device embolization [15]. Primary safety events were more frequent in the intervention group (7.4 per 100 patient-years [95% CrI: 5.5–9.7] vs 4.4 per 100 patient-years [95% CrI: 2.5–6.7]; relative risk: 1.69; 95% CrI: 1.01–3.19). The procedural success rate was 91%, and adverse safety events were mainly a result of periprocedural complications. Major procedural adverse events occurred among 12% of patients, mainly in the form of pericardial effusions (5%) necessitating percutaneous or surgical drainage and acute ischemic stroke (1%) due to air or thromboembolism. Four patients had to have the device removed, three because of device embolization (one removed percutaneously using a vascular snare and two surgically) and one because of postimplantation sepsis. It was obvious that the incidence of severe pericardial effusion was related to the center and operator experience, being 50% higher at newly initiated trial sites. Other serious device-related events included a 34% higher relative risk of ischemic stroke than the warfarin group, perhaps due to incomplete LAA closure or device-related thrombus (occurred in ∼4% of patients). Notably, of the 20 device-related thrombus findings, only three led to ischemic strokes (i.e., thrombus-associated annualized stroke rate of 0.3 per 100 patient-years). The remaining patients stayed asymptomatic with endothelialization or lysis of the thrombi under prolonged or resumed oral anticoagulation (oral anticoagulation for 6 weeks after the intervention was part of the protocol) [17].

Drawbacks of current generation devices & directions for improvement

There remain deficiencies with current technologies such as cardiac perforations, which may result from transseptal puncture, which will remain an integral part of most procedures (save patients with PFO or ASD). The respective risk greatly depends on anatomical variations and operator experience. Selective intubation of the LAA with catheter or delivery sheath represents a further risk for cardiac perforations. The soft-tipped, double-curved ACP delivery sheath may be the least traumatic and most user-friendly approach. Late pericardial effusions have been described probably due to erosion of the LAA. Device embolization has been effectively reduced with device modifications and active fixation (barbs for WATCHMAN and hooks for ACP). Malsizing or malpositioning due to the wide variation in sizes and particular shapes of LAA continue to occur as a risk factor for embolization. They are also a reason for incidentally observed incomplete LAA obliterations after device implantation, which however seem to be a rather benign finding not associated with increased stroke risk [18]. Concerns about endocardial metal devices have driven the investigation of other non-metal device occluders and an epicardial LAA exclusion approach. The SIDERIS transcatheter patch, a frameless balloon-deliverable device tailored from polyurethane foam, in conjunction with surgical adhesives was shown to be safe and feasible for endocardial LAA occlusion in a small study [16]. The Place procedure with the Lariat suture delivery device (Sentre Heart, CA, USA), applied epicardially via a standard percutaneous pericardial access and directed toward a target magnet wire placed in the LAA via a transseptal approach, demonstrated promising initial data among 13 patients, particularly in the sense of complete LAA occlusion, without leaving intracardiac material behind [19].

Conclusion

Although the PROTECT AF trial demonstrated that LAA closure was equally effective to warfarin, current clinical indications remain restricted to patients experiencing complications of or contraindications to oral anticoagulants, or those refusing their intake. This is mainly due to safety issues of current devices. Improvements in devices and techniques for transcatheter LAA occlusion are mandated, as are further randomized trials to allow widespread clinical application of this promising technique as a valid alternative for (or complement to) stroke prevention among AF patients with oral anticoagulation.

Financial & competing interests disclosure

AA Khattab has received proctor fees from St. Jude Medical, and B Meier has received research grants and proctor/speaker fees from St. Jude Medical. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.