Left Atrial Appendage Closure With Amplatzer Septal Occluder in Patients With Atrial Fibrillation: CT-Based Morphologic Considerations

Abstract: Objectives. This study analyzed left atrial appendage (LAA) dimensions measured by computed tomography (CT) scan to define optimal selection of Amplatzer septal occluders for LAA closure. Background. Patients with atrial fibrillation and contraindications to anticoagulation have limited options for LAA closure until approval of dedicated closure devices. Off-label use of available cardiac devices represents one option. Methods. All consecutive patients undergoing LAA occlusion with an Amplatzer device who had undergone CT scanning were included. Numerous dimensions of the LAA were measured in order to optimally select a device that would simultaneously provide good anchoring and good sealing of the LAA. Results. Of 11 eligible patients, 8 had successful Amplatzer deployment. In all successful cases, the putative “left atrial” disc was well matched to the “landing zone” of the LAA, proving good anchoring. The proximal (putative “right atrial”) disc was sized to cover the LAA orifice. The failed cases shed light on procedural variables. Conclusions. LAA occlusion with an Amplatzer device is a viable option for patients with atrial fibrillation and contraindications to anticoagulation. Careful attention to LAA dimensions as measured on CT scan assists in optimizing device selection.

J INVASIVE CARDIOL 2015;27(5):258-262

Key words: atrial fibrillation, CT scan, Amplatzer septal occluder

________________________________________________

While systemic anticoagulation with warfarin or one of the newer oral anticoagulants in patients with atrial fibrillation (AF) has been convincingly demonstrated to confer substantial risk reduction for stroke, a substantial portion of patients either have contraindications or other obstacles to anticoagulation and are thus unprotected against stroke risk. With the recognition that most demonstrable thrombi form within the left atrial appendage (LAA), mechanical approaches to close or obliterate the left atrial appendage are being developed as alternatives.^1-6

Initial clinical data with at least one dedicated device (Watchman; Aritech, Inc) show promise. At the time of this study, the device was not yet approved for clinical use in the United States, but a Food and Drug Administration (FDA) panel had provided a favorable review;¹ the device has since obtained FDA approval. Other devices specifically designed for occluding the LAA are also being evaluated, but their approval and commercialization will undoubtedly take several years.^2-5 The Lariat suture delivery device (SentreHeart, Inc), which is not specifically approved for LAA occlusion but is an otherwise approved, marketed product, is gaining popularity for “off-label” use.⁶ Controlled clinical trial data for this device are not available, and deploying the device necessitates an intrapericardial approach. Thus, patients with non-accessible pericardia (such as those with prior cardiac surgery) are ineligible for this procedure. Consequently, the search for alternatives continues.

LAA occlusion with cardiac closure devices designed for septal defects is an alternative that has received some attention for at least a decade.^7-10 These devices have the advantage of FDA approval and are thus readily available. Although their use for LAA occlusion represents off-label use, their familiarity to cardiologists and known safety profile make them a potentially attractive alternative until devices specifically engineered for the LAA become available. The literature on the use of these devices is modest.^7-10 There has not previously been any published data on the use of advanced imaging to understand the geometric aspects of device selection and insertion. The aim of this paper is to describe computed tomography (CT)-scan based analysis of the LAA geometry in employing Amplatzer septal closure devices (St. Jude Medical) to accomplish LAA occlusion.

Methods

All consecutive patients undergoing attempted LAA occlusion at Henry Ford Hospital were considered for analysis. With one exception, all patients had cardiac CT scans (with contrast) available for analysis. Patients were referred for Amplatzer LAA occlusion on the basis of clinical contraindications or obstacles to systemic anticoagulation in the setting of AF combined with added stroke risk factors (CHADS₂ or CHA₂DS₂-VASc). 

LAA morphology has been previously described as “windsock,” “chicken-wing,” “cactus,” or “cauliflower.”¹¹ For the goal of deploying an Amplatzer septal closure device in the LAA, the principal consideration is the presence of a circumferential rim of tissue beyond the neck of the LAA in order to anchor the device. Thus, patients with wind-sock LAA morphology were deemed ineligible for LAA closure with the Amplatzer septal closure device.

All procedures were performed percutaneously via a transfemoral venous approach with a transseptal puncture to access the left atrium (LA). Intraprocedural transesophageal echocardiography (TEE) was used to guide the transseptal puncture and placement of the device in the LAA. A pigtail catheter was positioned into the LAA. A relatively acute angle was formed on a Rosen wire, which was then advanced via the pigtail into the LAA; the pigtail was then withdrawn. A TorqVue device (St. Jude Medical), sized appropriately to the selected Amplatzer device, was advanced into the LAA with the dilator removed and the wire still in the LAA. Bleed-back was obtained, and then the Rosen was removed. The Amplatzer device was then advanced into the LAA and a small portion of the disc was advanced outside the sheath until it took the shape of a bulb. At this point, the device was pushed snugly into the LA; with firm forward pressure, the left disc was released. Finally, with continued firm pressure, the right atrial disc was released so that it was expanded adjacent to the LAA ostium. Confirmation that the device was firmly situated within the LAA was confirmed by a firm tug on the system (Figure 2; Video 1), after which the delivery cable was unscrewed from the device, releasing the device.

The measured parameters were selected on the basis of several principles. First, the nominal left atrial disc of the Amplatzer has to both fit within the LAA, as well as serve as the anchor to prevent embolization of the device from the LAA. Second, the nominal right atrial disc has to seat within or near the ostium of the LAA in a manner that sufficiently occludes the LAA and yet is as flush as possible with the contour of the neighboring body of the LA in order not to have a bulky device protruding into the LA. 

The measured parameters included minimum diameter (Dmin) as the narrowest diameter of the isthmus of the neck of the LAA. Although this parameter was not used in any substantial decision-making, it served as a reference point for assessing the other parameters. Maximum diameter (Dmax) was the longest chord that connected the farthest points of the LAA beyond the isthmus. The importance of Dmax in considering device selection is that the “left atrial” disc of the Amplatzer needs to be accommodated within this Dmax. An oversized non-fully deployed left atrial disc may still fit snugly within the LAA and be safe from embolization, albeit in distorted geometric form. Such a distorted geometry and non-deployed bulk (sometimes described as a “cobra head” rather than a discoidal shape) may prevent adequate seating of the right atrial disc in the desired occlusive and flush manner. The two most important parameters measured are labeled “D1” and “D2.” These represent the anchoring planes for the two discs, respectively. D1 and D2 were measured as follows. At the plane of Dmin, we called the walls of the LAA isthmus “vertical” purely as a point of reference. In either direction (that is, “inward” into the LAA and “outward” toward the body of the LA), the walls of the neck of the LAA splay to a less vertical, more “horizontal” angulation. We measured D1 at the points where the angles of the wall could be said to be truly horizontal (coinciding with the inflection point at which the angle of the wall ceases to splay away from the vertical and begins to return toward the vertical). In the direction of the body of the LA, however, there usually was no such inflection point, as the walls could remain at a fixed angle to the vertical, could pass the right-angle, or could continue in a nearly horizontal line indefinitely. D2 was thus selected at the point at which the walls were maximally deviated from the vertical. Line “L” connects the centers of D1 and D2, and assuming that the left and right atrial discs are well anchored at D1 and D2, represents the dimension within which the device waist will reside.

The ideal device selection would thus match device dimensions to the measured parameters. The LA disc would ideally be several millimeters larger than D1 and no larger than Dmax. The RA disc would be several millimeters larger than D2. The device waist length would approximate L. 

CT images were manipulated manually in three dimensions to optimally demonstrate the parameters needed. No single planar image could necessarily demonstrate all the parameters, although an effort was made to select a single image from which a maximum of the selected parameters could be measured.

Results

Of 12 patients undergoing LAA occlusion with an Amplatzer device, 11 are included in this report. One patient did not undergo preprocedural CT scanning and is thus not included in this analysis. Patient baseline clinical characteristics are summarized in Table 1. 

An Amplatzer device was successfully deployed in 9 out of 12 total patients. In these 9 patients, the device selected conformed to the CT-measured dimensions as described in the Methods section (Table 2). In 3 patients, attempts at deployment were abandoned, based largely on an inability to “unfurl” the distal disc in a satisfactory manner that would provide adequate anchoring of the device. The principal limiting issue was the inability to align the delivery of the device coaxially with the axis designated “L” in Figure 1. The root cause was a septal puncture that was relatively high, such that the approach to the LAA was close to perpendicular to axis L. The AGA TorqVue catheter does not permit any active angulation of the catheter that would obviate this issue. In the successful deployment cases, the septal puncture, as guided by TEE, was typically more inferior and posterior, thus permitting a more coaxial approach to the LAA.

In 1 patient with initial deployment, the device was noted to have dislodged on day 1 post procedure. During the deployment procedure, it was noted by three-dimensional TEE that the distal disc was not fully anchored within the intended landing zone. That is, part of the disc was above “line D1,” and part was beneath it. Because more than 50% of the rim of the distal disc was above “line D1,” it was initially thought to be sufficiently anchored. Dislodgment was identified by routine postprocedural echocardiographic surveillance, without any attendant clinical developments. The device was successfully snared from the left atrium (within which it had been freely mobile) via a transseptal approach, and deployed within the atrial septum with no resulting clinical sequelae. 

Patients were most commonly discharged on day 1 post procedure. The postdischarge antithrombotic regimen was aspirin 81 mg daily plus clopidogrel 75 mg daily. If possible, the clopidogrel was to be continued for 6 months and the aspirin indefinitely.

Patients have been contacted for follow-up. One individual could not be located; internet-based searches did not yield any evidence of death. Mean duration of postprocedure follow-up was 459 ± 233 days. Two patients had died of non-cardiac causes (urosepsis and myelodysplastic syndrome; acute respiratory distress syndrome). All other patients remain free of any embolic events or device-related complications.

Discussion

Stroke prevention in patients with AF and added stroke risk factors is of paramount importance, but in many cases contraindications to anticoagulation exist, prompting the pursuit of alternative strategies. Occlusion of the LAA, the presumed source of most emboli in patients with AF and stroke, is the target of devices in development as well as off-label use of marketed devices. We describe a series of patients in whom we pursued occlusion of the LAA with Amplatzer septal occluder devices because all other options were closed. Extracardiac suturing of the LAA with off-label use of a Lariat suture delivery device was not feasible because of either geometric issues with the LAA or inaccessibility of the pericardial space due to prior cardiac surgery.

In our analysis of the LAA geometry as assessed by cardiac CT, we defined the elements that optimize Amplatzer device selection. The device’s distal disc (the putative left atrial disc) needs to be carefully matched to the landing zone beyond the isthmus of the LAA. Similarly, the proximal (putative right atrial) disc needs to be larger than the ostium of the LAA. In review of the cases in which we failed to deliver the device, we offer the observation that in performing TEE-guided transseptal puncture, the preferable puncture site should be relatively inferior and posterior, or the angle of approach to the LAA will be too far to the perpendicular to permit adequate unfolding of the distal disc within the landing zone.

Finally, our single case of device dislodgment led us to conclude that the entire distal disc should be within the landing zone above the isthmus, and that any deviation from this may pose a risk of inadequate anchoring of the device.

The efficacy of LAA closure in stroke prevention, either with dedicated devices¹² or other off-label approaches,¹³ is as yet questioned by some commentators.^14,15 Our modest sample size precludes any definitive statements about long-term efficacy or safety, but the results thus far are reassuring. Issues of concern, such as late device embolization¹⁶ or erosion, have not occurred. 

Conclusion

Occlusion of the LAA with the Amplatzer septal closure device is a viable approach to patients with AF who are not candidates for anticoagulation or alternative LAA occlusion approaches until such time as dedicated devices have been approved for clinical use.

References

1. Reddy VY, Doshi SK, Sievert H, et al; PROTECT AF Investigators. Percutaneous left atrial appendage closure for stroke prophylaxis in patients with atrial fibrillation: 2.3-year follow-up of the PROTECT AF (Watchman Left Atrial Appendage System for Embolic Protection in Patients with Atrial Fibrillation) trial. Circulation. 2013;127:720-729. 
2. Lam YY. A new left atrial appendage occluder (Lifetech LAmbre device) for stroke prevention in atrial fibrillation. Cardiovasc Revasc Med. 2013;14:134-136. 
3. Urena M, Rodés-Cabau J, Freixa X, et al. Percutaneous left atrial appendage closure with the Amplatzer cardiac plug device in patients with nonvalvular atrial fibrillation and contraindications to anticoagulation therapy. J Am Coll Cardiol. 2013;62:96-102. 
4. Park JW, Bethencourt A, Sievert H, et al. Left atrial appendage closure with Amplatzer cardiac plug in atrial fibrillation: initial European experience. Catheter Cardiovasc Interv. 2011;77:700-706.
5. Rodés-Cabau J, Champagne J, Bernier M. Transcatheter closure of the left atrial appendage: initial experience with the Amplatzer cardiac plug device. Catheter Cardiovasc Interv. 2010;76:186-192.
6. Stone D, Byrne T, Pershad A. Early results with the Lariat device for left atrial appendage exclusion in patients with atrial fibrillation at high risk for stroke and anticoagulation. Catheter Cardiovasc Interv. 2013 Jun 13 (Epub ahead of print). 
7. Cruz-Gonzalez I, Cubeddu RJ, Sanchez-Ledesma M, et al. Left atrial appendage exclusion using an Amplatzer device. Int J Cardiol. 2009;134:e1-e3.
8. Meier B, Palacios I, Windecker S, et al. Transcatheter left atrial appendage occlusion with Amplatzer devices to obviate anticoagulation in patients with atrial fibrillation. Catheter Cardiovasc Interv. 2003;60:417-422.
9. Nietlispach F, Gloekler S, Krause R, et al. Amplatzer left atrial appendage occlusion: single-center 10-year experience. Catheter Cardiovasc Interv. 2013;82:283-289.
10. Aryana A, Bailey S, Gearoid O’Neill P, d’Avila A. Endocardial occlusion of incompletely surgically ligated left atrial appendage using an Amplatzer septal occluder device. Heart Rhythm. 2014;11:735-736. Epub 2013 Jun 20.
11. Di Biase L, Santangeli P, Anselmino M, et al. Does the left atrial appendage morphology correlate with the risk of stroke in patients with atrial fibrillation? Results from a multicenter study. J Am Coll Cardiol. 2012;60:531-538.
12. Holmes DR Jr, Kar S, Price MJ, et al. Prospective randomized evaluation of the Watchman left atrial appendage closure device in patients with atrial fibrillation versus long-term warfarin therapy: the PREVAIL trial. J Am Coll Cardiol. 2014;64:1-12.
13. Price MJ, Gibson DN, Yakubov SJ, et al. Early safety and efficacy of percutaneous left atrial appendage suture ligation: results from the US Transcatheter LAA Ligation Consortium. J Am Coll Cardiol. 2014;64:565-572.
14. Lee RJ. Evolution of stroke prevention in nonvalvular atrial fibrillation patients. J Am Coll Cardiol. 2014;64:13-15.
15. Dagres N, Rolf S, Hindricks G. Percutaneous left atrial appendage suture ligation: not ready for prime time. J Am Coll Cardiol. 2014;64:573-575.
16. Pérez Matos AJ, Swaans MJ, Rensing BJ, et al. Embolization of a left atrial appendage closure device unmasked by intermittent left bundle branch block. JACC Cardiovasc Interv. 2014;7:e115-e117.

_________________________________________

From ¹Cardiology Division, Henry Ford Hospital, Detroit, Michigan; and ²Bay Pines Veterans Affairs Health System, Bay Pines, Florida. 

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. The authors report no conflicts of interest regarding the content herein.

Manuscript submitted May 20, 2014, provisional acceptance given July 23, 2014, final version accepted October 28, 2014.

Address for correspondence: Paul T. Vaitkus, MD, Cardiology Division, Mailstop 111,

Bay Pines VA Medical Center, 10000 Bay Pines Blvd, Bay Pines, FL 33744. Email: paultvaitkus@yahoo.com