- Email: [email protected]
Early safety of the Amplatzer Cardiac Plug™ for left atrial appendage occlusion
International Journal of Cardiology 168 (2013) 3920–3925
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Early safety of the Amplatzer Cardiac Plug™ for left atrial appendage occlusion☆ David Meerkin a,⁎,1, Adi Butnaru b, Dmitry Dratva c, Olivier F. Bertrand d,e,f,g, Dan Tzivoni h,i a
Structural and Congenital Heart Disease Unit, Shaare Zedek Medical Center, Jerusalem, Israel Department of Cardiology, Shaare Zedek Medical Center, Jerusalem, Israel c Department of Cardiology, Shaare Zedek Medical Center, Jerusalem, Israel d Interventional Cardiologist, Quebec Heart–Lung Institute, Quebec City, Quebec, Canada e Faculty of Medicine, Laval University, Quebec City, Quebec, Canada f Department of Mechanical Engineering, McGill University, Quebec City, Quebec, Canada g Quebec Foundation for Health Research, Quebec City, Quebec, Canada h Department of Cardiology, Shaare Zedek Medical Center, Jerusalem, Israel i Medicine (Cardiology), Hebrew University, Jerusalem, Israel b
a r t i c l e
i n f o
Article history: Received 24 May 2012 Received in revised form 16 May 2013 Accepted 29 June 2013 Available online 23 July 2013 Keywords: Atrial fibrillation Left atrial appendage Thromboembolism Stroke Device occlusion
a b s t r a c t Objective: To assess the cumulative experience of a single operator using a strict set of deployment and release criteria for the Amplatzer Cardiac Plug™ (ACP) and the impact of these criteria on procedural success and complications. Background: Following strong evidence that the left atrial appendage (LAA) is the site of the majority of thrombus formation within the left atrium in patients with non-valvular atrial fibrillation, non-pharmacological approaches to LAA exclusion have been developed and shown to be effective. Methods: Procedural and in-hospital outcomes of LAA occlusion performed by or under the supervision of a single operator using the ACP™ in 100 anticoagulant ineligible patients with a high stroke risk were assessed. Results: One hundred patients with a mean CHADS2 score of 3.21 ± 1.23 underwent catheterization for closure of LAA with the ACP™. The mean landing zone as assessed by TEE was 20.01 ± 3.21 mm, and 20.8 ± 3.19 mm by fluoroscopy. The mean difference between the TEE and the fluoroscopic measurements was 0.8 ± 1.13 mm. Device deployment was successful in 100/100 attempted cases with a mean deployed device size of 24.36 ± 3.27 mm. Procedural complications were limited to a single case of pericardial tamponade and one postprocedural pulmonary edema both of which were adequately treated with no long-term sequelae. Conclusions: In this single operator report, LAA occlusion using the double element ACP™ can be safely performed with excellent success rates. Using very specific deployment success, stability and release criteria, this device can achieve LAA occlusion with extremely low complication rates in an extremely frail oral anticoagulant ineligible population with multiple co-morbidities. © 2013 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The most significant clinical implication of atrial fibrillation is its potential for left atrial thrombus formation with subsequent embolic phenomena, the most devastating of which is stroke. In patients with significant risk factors for stroke, the treatment of choice is therapeutic anticoagulation with Vitamin K antagonists [1–3]. The inconvenience and increased risk profile when the treatment strays from the therapeutic window has been a great stimulus for the development of alternative therapeutic strategies. Most recently the benefits of new classes of ☆ All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. ⁎ Corresponding author at: Structural and Congenital Heart Disease Unit, Shaare Zedek Medical Center, POB 3235, Jerusalem, 91031, Israel. Tel.: +972 2 655 5975; fax: +972 2 655 5437. E-mail address: [email protected] (D. Meerkin). 1 David Meerkin has the following conflict of interest: Proctor for St. Jude Medical. 0167-5273/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijcard.2013.06.062
drugs, particularly Factor Xa and direct thrombin inhibitors, have been demonstrated to be effective at reducing events in these patients with equivalent or improved safety profiles [4–6]. In spite of their benefits, there remains a significant population of untreated patients, predominantly with relative or absolute contraindications to most forms of anticoagulants [7,8]. Following strong evidence that the left atrial appendage (LAA) is the site of the majority of thrombus formation within the left atrium in patients with non-valvular atrial fibrillation [9], non-pharmacological approaches to LAA exclusion have been developed. Initial experiences of transcatheter LAA occlusion have been published with three distinct dedicated devices, PLAATO (currently unavailable), Watchman (Atritech Inc., Minneapolis MN) and the Amplatzer Cardiac Plug™ (ACP) (AGA Medical Inc., Minneapolis MN) [10–14]. This experience, as presented with all three devices, demonstrated a relatively high rate of in-hospital safety events, particularly pericardial effusion and tamponade, peri-procedural strokes, air embolism and device embolization requiring device snaring and
D. Meerkin et al. / International Journal of Cardiology 168 (2013) 3920–3925
removal or surgical intervention, as well as failure to successfully deploy. Most recently, Reddy et al. published the results of the Continued Access Protocol (CAP) of the PROTECT AF trial using the WATCHMAN device [15]. In this registry, centers enrolling patients in the PROTECT AF trial had the right to continue to treat patients with the same inclusion and exclusion criteria with the study device. The authors emphasized the findings of significant reduction of in hospital complications following the generation of greater operator experience. This was evident by the reduction of complications when CAP enrolled patients were compared with those of the initial PROTECT AF patients. Park and colleagues reported the initial European experience with the ACP™ [14]. This series of 137 patients treated at 10 centers across Europe represented the operators' initial experience with the device and as such included a variety of deployment techniques (including the first 11 cases treated by these authors). In contrast, we report here the cumulative experience of a single operator using a strict set of deployment and release criteria for the ACP™ and the impact of these criteria on procedural success and complications. 2. Methods 2.1. The Amplatzer Cardiac Plug left atrial appendage occlusion procedure 2.1.1. Device description The ACP™ device (AGA Medical Inc., Minneapolis MN), procedure and acute results have been previously described [14]. It is formally approved and commercially available in many countries including most of Europe, Australia, Canada and the Middle East. The system consists of a delivery catheter, a deployment cable and the plug itself. The threedimensional shaped delivery catheter is positioned in the LAA following trans-septal puncture. The ACP™ itself is made of nitinol wire mesh and polyester patch shaped in a two-part configuration. The distal part is shaped much as a hockey puck and is termed the device lobe. This connects to a more proximal and larger disk with a small connecting waist. Small nitinol retainers protrude from the expanded lobe to engage the tissue of the appendage at the site of deployment to prevent dislodgment in the direction of the LA. The retainers are retracted and unexposed when the device is collapsed and elongated or inadequately expanded. The device is connected to a delivery cable by the standard Amplatzer screw attachment. The device is sized by the lobe diameter and is available in the range of 16–30 mm in 2 mm increments. The disk is 4 mm larger than the lobe in sizes b24 mm and 6 mm larger in sizes ≥24 mm. 2.1.2. Implantation description The procedure was performed under general anesthetic and transesophageal echocardiographic (TEE) guidance. The majority of patients underwent preparatory TEE the day prior to the procedure to exclude LAA thrombus and for initial assessment of size and morphology. Intraprocedurally TEE sizing was repeated. At least four TEE views were attained for LAA assessment. As the LAA ostium and neck are elliptical in shape, the 0 or 180° provide a measure of their long axes. Measures were then performed in the 45 and 90° orientations. In these views, the pulmonary vein ridge and circumflex artery were identified and a connecting line drawn between them representing the LAA ostium. As the device lobe is intended to sit at a depth of 1 cm within the LAA neck, a line was drawn from the middle of the initial ostial measuring line along the axis of the LAA neck (not its body). At approximately 10–12 mm along this line a second line was measured perpendicular to the LAA neck axis from a point beneath the circumflex artery to the opposing wall (Fig. 1). This site is termed the landing zone and its measure should correlate with an equivalent site on fluoroscopic measure (Fig. 2). Selection of device size was based upon the larger of the available measures. Devices were available to treat landing zones measuring 12.5–28.5 mm. Finally the LAA was interrogated in the 120–140° to assess distal lobes and further clarify the appendage anatomy. Using a transradial or transfemoral artery approach a 5F marker pigtail (Cordis, Miami Lakes, FL) was placed at the site of the aortic valve. This assisted as a marker for site positioning the transeptal puncture, calibration during LAA contrast injection and monitoring of aortic pressures during the procedure. Calibration was performed using the 10 mm marker distance on a straight segment of the marker pigtail close to the LAA. Following attainment of right or left femoral venous access, transeptal puncture with a standard Mullins Sheath (Medtronic, Minneapolis, MN) and Brockenbrough needle (8F, Medtronic, Minneapolis, MN) was performed under TEE guidance with the site of puncture being aimed at the inferior portion of the membranous fossa ovalis, mid to posteriorly located. Use of alternating bicaval and short axis TEE views allowed for precise positioning. This puncture site facilitates subsequent orientation of the delivery system in the appendage and adequate space to maneuver the guide within the left atrium. In the three of first 10 patients, access to the LA was through a PFO. Subsequently, access to the LA was always through a transeptal puncture to improve catheter manipulation and orientation in the LA and LAA.
3921
A 5F multipurpose catheter with sideholes (Cordis, Miami Lakes, FL) was advanced though the Mullins sheath and using a combination of the angulations of the sheath and catheter under TEE and fluoroscopic guidance, maneuvered into the LAA. Selective contrast injections were performed in four standard views to interrogate the LAA. These views are AP, RAO 30°, RAO 30°/Cranial 15° and RAO 30°/Caudal 15°. In selected patients with impaired renal function, this was restricted to a single (RAO 30°/Cranial 15°) or occasionally two views. Quantitative angiographic analysis (Siemens Artis software, Munich, Germany) was performed in the best 2–3 views that demonstrated the LAA neck. A single frame demonstrating maximal appendage diameters was selected. A line was drawn across the LAA ostium. A second 10 mm line originating from the middle of the ostium and running in the axis of the LAA neck was then drawn. The measure of the landing zone was a third line perpendicular to the LAA neck axis running from the device engagement site beneath the circumflex coronary artery to the opposing wall (Fig. 2). Correlation of fluoroscopic sizing with the TEE measures within 2 mm was required prior to device selection. If a discrepancy between modality measures was noted, the measures were repeated including recalibration. Target correlation was achieved in all cases. The Amplatzer 45° × 45° TorqVue™ 3-dimensional delivery sheath (TorqueVue, AGA, Minneapolis MN) was used in almost all cases (97/100 due to availability). Sheath caliber selection was based upon device size selection according to company recommendations. Exchange was performed with a stiff 0.035″ guidewire (002 Amplatzer wire AGA, Minneapolis MN) positioned alternatively in the LAA or left upper pulmonary vein (LUPV). Once the delivery sheath was positioned in the LAA at a depth of approximately 2 cm, the device was advanced forward to the sheath tip. Under TEE guidance the initial portion of the device was unsheathed allowing it to attain a ball shape, excluding the first 20 patients. In those patients the device was advanced from within the sheath. At approximately the midpoint of expansion (designated by radio opaque markers), the sheath was stabilized and the device advanced. This allowed the device to return to its intended shape (much like a car tire), without advancing forward and as such not risking perforation. The device lobe position was then assessed by echo and contrast fluoroscopy and if the position was satisfactory the device disk was unsheathed within the LA and when exposed allowed to retract to its position covering the LAA ostium (Fig. 3). Due to extreme anatomical variability, device lobe positioning was not always as anticipated. If the operator interpreted that the lobe position was off angle, too deep or too shallow or if the disk position was considered too deep or disruptive to other structures, the device was resheathed and redeployed in a more favorable position. With position assessed by echo and contrast fluoroscopy, five criteria (which have subsequently been included in the company IFU) were used to confirm stability and adequate position: 1. Adequate lobe compression such that it resembles a car tire (and not a hockey puck as when unrestrained) 2. Lobe orientation being perpendicular to the LAA neck axis 3. Adequate separation of the device lobe and disk (as compared to the unrestrained position) 4. Device disk concavity, representing tension from the device lobe transferred though the connecting waist and improved likelihood of disk occlusion of the LAA ostium 5. Two thirds of the device lobe was located distal to the circumflex artery as visualized by TEE. If it was clear that the device size was inappropriate the device was retrieved, sheath changed and an alternate device deployed. When the above five criteria were achieved by both TEE and fluoroscopic imaging the device was released by unscrewing the delivery cable. Initially a “tug test” was performed to confirm stability prior to release. This was abandoned after the first 15 patients and the device released once the five release criteria were confirmed. Repeat transthoracic echocardiography was performed the next day to confirm device position and exclude embolization. In the event that the device could not be clearly discerned by echocardiography, fluoroscopy was performed to exclude embolization.
Fig. 1. Echocardiographic measurement of left atrial appendage (LAA) for device sizing at device lobe landing zone (dashed line) at 1 cm depth (solid line) from the appendage ostium (dotted line).
3922
D. Meerkin et al. / International Journal of Cardiology 168 (2013) 3920–3925 2.4. Statistical analysis Events are expressed as absolute numbers or percentages. Continuous variables are expressed as mean ± standard deviation or ranges. The authors of this manuscript have certified that they comply with the Principles of Ethical Publishing in the International Journal of Cardiology.
3. Results 3.1. Patient demographics
Fig. 2. Fluoroscopic measurement of left atrial appendage (LAA) for device sizing at device lobe landing zone (dashed line) at 1 cm depth from the appendage ostium (dotted line). This image was taken at an angulation of RAO 30°, Cranial 15°. PVR = Pulmonary vein ridge.
2.2. Anticoagulation All patients were treated with a standard heparin bolus dose of 5000–10,000 units administered after transeptal puncture with additional boluses to maintain an ACT of N250 s. All patients were treated with a loading dose of 300 mg of aspirin and 300 mg of clopidogrel post-procedure followed by a lifelong recommendation of 100 mg aspirin daily with 75 mg of clopidogrel recommended to be continued for one month only. Major adverse events were defined as death, stroke, systemic embolism, device embolization, pericardial bleeding requiring an intervention (cardiac tamponade) or other major bleeding requiring invasive treatment or blood transfusion occurring during index hospital admission. All procedures were performed by or under the direct supervision of a single operator (DM) and were performed in the absence of any formal surgical back-up. 2.3. Patient population The procedure was offered as a clinical indicated solution to patients with paroxysmal, persistent or permanent atrial fibrillation, a CHADS2 score of ≥2 and a contraindication or inability to be treated with Vitamin K antagonists. Anonymous patient and procedural information were included in a clinical database. This study is the retrospective analysis of consecutive patients treated by a single operator included in the clinical database. Forty-two cases were performed in a single center (SZMC, Jerusalem, Israel) and 58 were performed as demonstration or training cases in 23 centers in Europe, Australia and India. In the training cases, in-hospital outcomes were confirmed via contact with the managing team either by direct phone contact or through the local device distributor. Patients with LAA thrombus or aneurysm were excluded. In the advent of an LAA thrombus as assessed by TEE, if possible, the patient was treated with low molecular weight heparin for 3–4 weeks and returned for repeat assessment and occlusion. All patients signed informed consent for the procedure. Patients included in a randomized clinical trial of the device were also excluded from the analysis.
Between June 2009 and March 2012, one hundred patients underwent catheterization for closure of left atrial appendage with the Amplatzer Cardiac Plug (AGA, Minneapolis, US). Patient demographics are shown in Table 1. There were 55 males. The mean age was 73 ± 9.95 years. There were 48 patients ≥75 and 83 ≥ 65 years of age. 27 patients were ≥80 years of age. 58 patients suffered from permanent while 16 had persistent and 26 paroxysmal atrial fibrillation. 53 patients had previously suffered an ischemic neurological episode. Diabetes mellitus was present in 25 and hypertension in 94 patients. The mean CHADS2 score was 3.21 ± 1.23. 3.2. Contraindications The range of contraindications for chronic anticoagulation was broad including both absolute and relative contraindications as shown in Table 2. The most frequent reason for referral was bleeding with and without anticoagulants and the most frequent site was from the gastrointestinal tract. Lack of compliance with either medication or INR measures was also not infrequent at 14%. Additional, unusual causes included bone pain, allergies, residence in a remote area without INR measurement capabilities. 3.3. Landing zone measurements and device deployment Measures of the landing zone by TEE and by angiographic techniques are shown in Table 3. The mean landing zone as assessed by TEE was 20.01 ± 3.21 mm, while the mean landing zone measured by fluoroscopy usually at RAO 30° and RAO 30/Cranial 15° was 20.8 ± 3.19 mm. The mean difference between the TEE and the fluoroscopic measurements were 0.8 ± 1.13 mm. Based upon these measurements the device to be deployed and the delivery sheath were selected. In all cases the landing zone measurements of the LAA were such that the device closure with the ACP™ could be attempted. Device deployment was successful in 100/100 attempted cases. The mean device size deployed was 24.34 ± 3.27 mm. The deployed device sizes are shown in Fig. 4. Interestingly, 4 device sizes (22, 24, 26 and 28 mm) comprised 80% of the deployed devices representing a size range of 6 mm. In 18 cases the initially selected device incorrectly sized and the device removed and replaced with a second device. In
Fig. 3. ACP device deployment and position in the left atrial appendage. ACP = Amplatzer Cardiac Plug, Lt. Cx = left circumflex, PV = pulmonary vein.
D. Meerkin et al. / International Journal of Cardiology 168 (2013) 3920–3925
all these cases the second device was successfully deployed. There were no instances when a third device was taken. In 14 of these 18 cases the device was upsized. This was by 2 mm in 12 cases and 4 mm in 2 cases. The device was downsized in only 4 cases, three by 2 mm and in another case by 6 mm. With increased experience this was less frequent with only 5 of these 18 cases being in the second half of the patient series. Radiation time, contrast volume and post-procedural hospital stay were available for the cohort of patients treated in the single center (SZMC) and are presented in Table 4.
Table 2 Contraindications to anticoagulation. Bleeding GI Intracranial Ocular Other sources (epistaxis/respiratory etc.) Compliance Falls Sundry
67 40 15 2 11 14 9 11
Values are presented as number of patients. GI = gastrointestinal.
3.4. Complications There were two significant in-hospital complications (Table 4). The first was a pericardial effusion with tamponade that occurred during device deployment in a shallow LAA in a 79 year-old woman a CHADS2 score of 3 and severe lower GI bleeding (patient number 20). This occurred as the device was being pushed out of the delivery sheath. The complication was noted immediately, the device deployed to exclude the leak from the left atrium and the effusion was drained urgently. The patient recuperated well and was discharged on hospitalization day 6 with no further complications. The second complication occurred in an 82 year-old woman with a CHADS2 score of 4 and inability to control INR. This woman additionally suffered from severely impaired left ventricular systolic function and severe mitral incompetence with mild renal failure. Following an uncomplicated procedure during which she received 80 cm3 of nonionic contrast material, extubation was performed as routine. One hour later she developed acute respiratory distress with pulmonary edema. She was reintubated and repeat TEE performed demonstrating the ACP well in position and worsening of the mitral incompetence probably due to additional volume load. Coronary angiography was performed to exclude compression of the circumflex artery and was unremarkable. The patient responded well to diuretics and was extubated 6 h later and discharged on hospital day 4. 4. Discussion Two devices have clearly demonstrated the benefits of transcatheter occlusion of the left atrial appendage, the PLAATO device in published registries [10–12] and the Watchman device in the Protect AF trial [13] and subsequently the CAP registry [15]. The PLAATO device, the
Table 1 Patient demographics. Age (years) Male gender CHADS2a score CHADS2 score (%) 0 1 2 3 4 5 6 Congestive heart failure Hypertension Age N75 years Diabetes mellitus Previous embolic neurological event Atrial fibrillation pattern Paroxysmal Persistent Permanent
3923
73 ± 9.95 55 3.21 ± 1.23 0 7 24 30 21 16 2 49 94 48 25 53 26 16 58
Values are presented as number of patients or mean ± SD. a Clinical prediction rule for estimating the risk of stroke in patients with non-rheumatic atrial fibrillation (AF).
first of its class, demonstrated a reduction of expected events in registries that included high risk patients that were ineligible for oral anticoagulants [10–12]. The principal limitation of the device use was its relative high incidence of periprocedural complications and MACCE. For commercial reasons this device did not progress beyond the initial studies arresting the follow up in the longer-term registry [10]. The Watchman device, in a carefully planned and executed randomized trial demonstrated the non-inferiority of the technique when compared with oral anticoagulants in eligible patients. As a result, the patient population was less sick and therefore at lower stroke risk than the populations studied with the PLAATO device. The limitation of the technique as was manifest in PROTECT AF study appeared to be the relatively high rate of periprocedural complications [13]. These complications included device embolization, periprocedural stroke and air embolism, however the most frequent complication was significant pericardial effusion/tamponade. When assessing any new therapy the risk benefit ratio is one of the cardinal relationships that must be determined. As this important trial was performed in 59 sites in the US and Europe, a very significant portion of these patients were treated within the learning curve of the center and as such were exposed to higher risks of intraprocedural complications. The recent publication of the Continued Access Registry result from the 26 highest volume centers included in the PROTECT AF trial demonstrated that with refinement of technique and increased experience, a significant reduction of procedural complications could be achieved with the Watchman device [15]. The initial European experience with the Amplatzer Cardiac Plug™ was published last year and demonstrated the feasibility of the LAA occlusion with this device [14]. The ACP acts mechanistically according to a different principal than its predecessors and in distinction to its name, does not attempt to plug the LAA. It functions by placing a retainer in the LAA neck and covering its ostium with a second disk shaped element much as a lid would to a box. The patient population in this early experience was high risk and predominantly anticoagulant ineligible. This population is at particularly high risk for embolic stroke, bleeding complications and intraprocedural complications. As the emphasis of that publication was on the technical aspects of the device, much of the patient baseline data was not reported. Additionally, the report included 137 patients treated at 10 sites, such that each center had limited experience, much as occurred in the PROTECT AF trial. The safety findings from this early experience reported to 24 h demonstrated an event rate of 7% including pericardial effusions requiring drainage (3.5%) and embolization (1.4%). This correlated, with that of the PROTECT AF, where a higher than desirable number of intraprocedural complications were reported, predominantly driven by pericardial effusions requiring drainage (4.8%, embolization 0.6%), resulting in a complication rate of 10.2%. As an extension to this initial experience we present our safety and feasibility results to further define the true risk of the procedure in experienced hands as well as the further definition of the high risk population that are candidates for this therapy in our clinical practice. Our population included patients at particularly high risk for atrial fibrillation associated embolic stroke with a mean CHADS2 score of 3.24 ± 1.20, with the vast majority of patients over the age of 65 and
3924
D. Meerkin et al. / International Journal of Cardiology 168 (2013) 3920–3925
Table 3 Procedural demographics. Flouro landing zone size (mm) TEE landing zone size (mm) Delta flouro-TEE ACP initial selection size (mm) ACP deployed size (mm) Guide size (F)
Table 4 Procedural outcomes. 20.8 20.01 0.8 24.24 24.36 11.89
± ± ± ± ± ±
3.19 3.21 1.13 3.23 3.27 1.5
Values are presented as mean ± SD. ACP = Amplatzer Cardiac Plug, TEE = transesophageal echo.
at least half of them post-previous stroke. Their frailty, although not measured in the CHADS2 score is more manifest by the causes of anticoagulant ineligibility, the leading cause being GI bleeding. In spite of an extremely high preprocedural risk for complications, our complication rates were low and the procedure was extremely well tolerated. In the Continued Access Registry of the Watchman device, improved technique and patient selection allowed for a 95% implant success rate. In our current report with the ACP™, there were no cases in which the device was found inappropriate for patient anatomy, being able to be deployed in all attempted cases. Exclusion of cases was not based upon LAA measures but rather, upon extremely poor patient condition and LAA thrombus. Theoretically however, extremely large (landing zone N 28.5 mm) and small (landing zone b 12.5 mm) measurements would have been excluded. A single case of LAA aneurysm (N6 cm) was excluded from the analysis. These encouraging results further justify the use of this procedure in this difficult population and the paucity of complications shifts the risk benefit interaction away from the status of a procedure with a high complication rate with associated morbidities. An alternative role for this procedure may also be as an interim measure for patients who temporarily require the discontinuation of oral anticoagulants due to bleeding or surgical interventions. As the safety profile improves the risk for the short-term benefit is reduced and it may present as an attractive alternative. In spite of the acceptance that the LAA occlusion is effective for embolic protection in atrial fibrillation as a class effect, long-term follow-up is required to confirm the level of efficacy of this device, with particular emphasis on device stability, thrombus formation, device leaks and stroke rates. These data will be necessary to allow for the further expansion of the indications of the device to patients who are candidates for long-term anticoagulants.
Successful deployment Deployment criteria Tire shape Separation Disk concavity Adequate orientation Lobe 2/3 behind Cxa (echo measurement) Residual flow (behind device lobe) TEEb color flow Doppler Fluoroscopic direct contrast injection Radiation time (min)c Contrast volume (cm3)c Procedural complications Death Stroke Myocardial infarction Air embolism Device embolism Pericardial effusion Post-procedural hospital stay (days)c Mean Median
100 99 100 100 100 100 0 4 15.4 ± 7.1 95.8 ± 42.0 0 0 0 0 0 1 2.4 ± 4.1 1
Values are presented as number of patients or mean ± SD. a Cx = circumflex artery. b TEE = transesophageal echo. c n = 42 patients treated in SZMC.
4.1. Limitations Occlusion of the left atrial appendage has inherent risks due to the required transeptal puncture and subsequent hardware manipulation within the appendage with potentially fragile appendage and extremely thin wall. A retrospective single operator study, although able to demonstrate the results achievable by an experienced operator, is limited in its implications to the broader penetration of the procedure. Our results however do include the learning curve, which seems to be principally manifest by errors in sizing and fewer occurrences of significant complications. This initial report furthermore only focuses on the acute results of device deployment and periprocedural complications. The true efficacy of the device must be measured by stroke prevention, residual leaks and device related thrombus in the longer term, in combination with these periprocedural results. However due to the presence of previous publications with the ACP™ and other devices describing relatively high early complication rates, we felt that reporting the results achieved with specific technical guidelines could potentially contribute to a significant reduction in complication rates achieved with the ACP™. Our long-term results specifically targeted at clinical and echocardiographic outcomes will be reported subsequently. 5. Conclusions In this single operator report, left atrial appendage occlusion using the double element ACP™ can be safely performed with excellent success rates. Using very specific deployment success, stability and release criteria, this device can achieve LAA occlusion with extremely low complication rates in an extremely frail oral anticoagulant ineligible population with multiple co-morbidities. The efficacy of this device needs to be assessed with longer-term clinical and echocardiographic follow-up. The low peri-procedural complication rate however, offers promise for the application of this device in the oral anticoagulant eligible population as well as the high risk, no-option population presented in this report. References
Fig. 4. ACP device size distribution range.
[1] Hart RG, Pearce LA, Aguilar MI. Meta-analysis: antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial fibrillation. Ann Intern Med 2007;146:857–67. [2] European Heart Rhythm Association, European Association for Cardio-Thoracic Surgery, Camm AJ, et al. Guidelines for the management of atrial fibrillation: the
D. Meerkin et al. / International Journal of Cardiology 168 (2013) 3920–3925
[3]
[4] [5] [6] [7] [8]
[9]
Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Eur Heart J 2010;31:2369–429. Fuster V, Rydén LE, Cannom DS, et al. ACCF/AHA/HRS focused updates incorporated into the ACC/AHA/ESC 2006 Guidelines for the management of patients with atrial fibrillation: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines developed in partnership with the European Society of Cardiology and in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. J Am Coll Cardiol 2011;57: e101–98. Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med 2009;361:1139–51. Patel MR, Mahaffey KW, Garg J, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med 2011;365:883–91. Granger CB, Alexander JH, McMurray JJ, et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2011;365:981–92. Bungard TJ, Ghali WA, Teo KK, McAlister FA, Tsuyuki RT. Why do patients with atrial fibrillation not receive warfarin? Arch Intern Med 2000;160:41–6. Nieuwlaat R, Capucci A, Camm AJ, et al. Atrial fibrillation management: a prospective survey in ESC member countries: the Euro Heart Survey on atrial fibrillation. Eur Heart J 2005;26:2422–34. Blackshear JL, Odell JA. Appendage obliteration to reduce stroke in cardiac surgical patients with atrial fibrillation. Ann Thorac Surg 1996;61:755–9.
3925
[10] Bayard YL, Omran H, Neuzil P, et al. PLAATO (Percutaneous Left Atrial Appendage Transcatheter Occlusion) for prevention of cardioembolic stroke in nonanticoagulation eligible atrial fibrillation patients: results from the European PLAATO study. EuroIntervention 2010;6:220–6. [11] Park JW, Leithäuser B, Gerk U, Vrsansky M, Jung F. Percutaneous left atrial appendage transcatheter occlusion (PLAATO) for stroke prevention in atrial fibrillation: 2-year outcomes. J Invasive Cardiol 2009;21:446–50. [12] Ostermayer SH, Reisman M, Kramer PH, et al. Percutaneous left atrial appendage transcatheter occlusion (PLAATO system) to prevent stroke in high-risk patients with non-rheumatic atrial fibrillation: results from the international multi-center feasibility trials. J Am Coll Cardiol 2005;46:9–14. [13] Holmes DR, Reddy VY, Turi ZG, et al. Percutaneous closure of the left atrial appendage versus warfarin therapy for prevention of stroke in patients with atrial fibrillation: a randomised non-inferiority trial. Lancet 2009;374:534–42. [14] 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;77:700–6. [15] Reddy VY, Holmes D, Doshi SK, Neuzil P, Kar S. Safety of percutaneous left atrial appendage closure: results from the Watchman Left Atrial Appendage System for Embolic Protection in Patients with AF (PROTECT AF) clinical trial and the Continued Access Registry. Circulation 2011;123:417–24.
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Early safety of the Amplatzer Cardiac Plug™ for left atrial appendage occlusion☆ David Meerkin a,⁎,1, Adi Butnaru b, Dmitry Dratva c, Olivier F. Bertrand d,e,f,g, Dan Tzivoni h,i a
Structural and Congenital Heart Disease Unit, Shaare Zedek Medical Center, Jerusalem, Israel Department of Cardiology, Shaare Zedek Medical Center, Jerusalem, Israel c Department of Cardiology, Shaare Zedek Medical Center, Jerusalem, Israel d Interventional Cardiologist, Quebec Heart–Lung Institute, Quebec City, Quebec, Canada e Faculty of Medicine, Laval University, Quebec City, Quebec, Canada f Department of Mechanical Engineering, McGill University, Quebec City, Quebec, Canada g Quebec Foundation for Health Research, Quebec City, Quebec, Canada h Department of Cardiology, Shaare Zedek Medical Center, Jerusalem, Israel i Medicine (Cardiology), Hebrew University, Jerusalem, Israel b
a r t i c l e
i n f o
Article history: Received 24 May 2012 Received in revised form 16 May 2013 Accepted 29 June 2013 Available online 23 July 2013 Keywords: Atrial fibrillation Left atrial appendage Thromboembolism Stroke Device occlusion
a b s t r a c t Objective: To assess the cumulative experience of a single operator using a strict set of deployment and release criteria for the Amplatzer Cardiac Plug™ (ACP) and the impact of these criteria on procedural success and complications. Background: Following strong evidence that the left atrial appendage (LAA) is the site of the majority of thrombus formation within the left atrium in patients with non-valvular atrial fibrillation, non-pharmacological approaches to LAA exclusion have been developed and shown to be effective. Methods: Procedural and in-hospital outcomes of LAA occlusion performed by or under the supervision of a single operator using the ACP™ in 100 anticoagulant ineligible patients with a high stroke risk were assessed. Results: One hundred patients with a mean CHADS2 score of 3.21 ± 1.23 underwent catheterization for closure of LAA with the ACP™. The mean landing zone as assessed by TEE was 20.01 ± 3.21 mm, and 20.8 ± 3.19 mm by fluoroscopy. The mean difference between the TEE and the fluoroscopic measurements was 0.8 ± 1.13 mm. Device deployment was successful in 100/100 attempted cases with a mean deployed device size of 24.36 ± 3.27 mm. Procedural complications were limited to a single case of pericardial tamponade and one postprocedural pulmonary edema both of which were adequately treated with no long-term sequelae. Conclusions: In this single operator report, LAA occlusion using the double element ACP™ can be safely performed with excellent success rates. Using very specific deployment success, stability and release criteria, this device can achieve LAA occlusion with extremely low complication rates in an extremely frail oral anticoagulant ineligible population with multiple co-morbidities. © 2013 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The most significant clinical implication of atrial fibrillation is its potential for left atrial thrombus formation with subsequent embolic phenomena, the most devastating of which is stroke. In patients with significant risk factors for stroke, the treatment of choice is therapeutic anticoagulation with Vitamin K antagonists [1–3]. The inconvenience and increased risk profile when the treatment strays from the therapeutic window has been a great stimulus for the development of alternative therapeutic strategies. Most recently the benefits of new classes of ☆ All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. ⁎ Corresponding author at: Structural and Congenital Heart Disease Unit, Shaare Zedek Medical Center, POB 3235, Jerusalem, 91031, Israel. Tel.: +972 2 655 5975; fax: +972 2 655 5437. E-mail address: [email protected] (D. Meerkin). 1 David Meerkin has the following conflict of interest: Proctor for St. Jude Medical. 0167-5273/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijcard.2013.06.062
drugs, particularly Factor Xa and direct thrombin inhibitors, have been demonstrated to be effective at reducing events in these patients with equivalent or improved safety profiles [4–6]. In spite of their benefits, there remains a significant population of untreated patients, predominantly with relative or absolute contraindications to most forms of anticoagulants [7,8]. Following strong evidence that the left atrial appendage (LAA) is the site of the majority of thrombus formation within the left atrium in patients with non-valvular atrial fibrillation [9], non-pharmacological approaches to LAA exclusion have been developed. Initial experiences of transcatheter LAA occlusion have been published with three distinct dedicated devices, PLAATO (currently unavailable), Watchman (Atritech Inc., Minneapolis MN) and the Amplatzer Cardiac Plug™ (ACP) (AGA Medical Inc., Minneapolis MN) [10–14]. This experience, as presented with all three devices, demonstrated a relatively high rate of in-hospital safety events, particularly pericardial effusion and tamponade, peri-procedural strokes, air embolism and device embolization requiring device snaring and
D. Meerkin et al. / International Journal of Cardiology 168 (2013) 3920–3925
removal or surgical intervention, as well as failure to successfully deploy. Most recently, Reddy et al. published the results of the Continued Access Protocol (CAP) of the PROTECT AF trial using the WATCHMAN device [15]. In this registry, centers enrolling patients in the PROTECT AF trial had the right to continue to treat patients with the same inclusion and exclusion criteria with the study device. The authors emphasized the findings of significant reduction of in hospital complications following the generation of greater operator experience. This was evident by the reduction of complications when CAP enrolled patients were compared with those of the initial PROTECT AF patients. Park and colleagues reported the initial European experience with the ACP™ [14]. This series of 137 patients treated at 10 centers across Europe represented the operators' initial experience with the device and as such included a variety of deployment techniques (including the first 11 cases treated by these authors). In contrast, we report here the cumulative experience of a single operator using a strict set of deployment and release criteria for the ACP™ and the impact of these criteria on procedural success and complications. 2. Methods 2.1. The Amplatzer Cardiac Plug left atrial appendage occlusion procedure 2.1.1. Device description The ACP™ device (AGA Medical Inc., Minneapolis MN), procedure and acute results have been previously described [14]. It is formally approved and commercially available in many countries including most of Europe, Australia, Canada and the Middle East. The system consists of a delivery catheter, a deployment cable and the plug itself. The threedimensional shaped delivery catheter is positioned in the LAA following trans-septal puncture. The ACP™ itself is made of nitinol wire mesh and polyester patch shaped in a two-part configuration. The distal part is shaped much as a hockey puck and is termed the device lobe. This connects to a more proximal and larger disk with a small connecting waist. Small nitinol retainers protrude from the expanded lobe to engage the tissue of the appendage at the site of deployment to prevent dislodgment in the direction of the LA. The retainers are retracted and unexposed when the device is collapsed and elongated or inadequately expanded. The device is connected to a delivery cable by the standard Amplatzer screw attachment. The device is sized by the lobe diameter and is available in the range of 16–30 mm in 2 mm increments. The disk is 4 mm larger than the lobe in sizes b24 mm and 6 mm larger in sizes ≥24 mm. 2.1.2. Implantation description The procedure was performed under general anesthetic and transesophageal echocardiographic (TEE) guidance. The majority of patients underwent preparatory TEE the day prior to the procedure to exclude LAA thrombus and for initial assessment of size and morphology. Intraprocedurally TEE sizing was repeated. At least four TEE views were attained for LAA assessment. As the LAA ostium and neck are elliptical in shape, the 0 or 180° provide a measure of their long axes. Measures were then performed in the 45 and 90° orientations. In these views, the pulmonary vein ridge and circumflex artery were identified and a connecting line drawn between them representing the LAA ostium. As the device lobe is intended to sit at a depth of 1 cm within the LAA neck, a line was drawn from the middle of the initial ostial measuring line along the axis of the LAA neck (not its body). At approximately 10–12 mm along this line a second line was measured perpendicular to the LAA neck axis from a point beneath the circumflex artery to the opposing wall (Fig. 1). This site is termed the landing zone and its measure should correlate with an equivalent site on fluoroscopic measure (Fig. 2). Selection of device size was based upon the larger of the available measures. Devices were available to treat landing zones measuring 12.5–28.5 mm. Finally the LAA was interrogated in the 120–140° to assess distal lobes and further clarify the appendage anatomy. Using a transradial or transfemoral artery approach a 5F marker pigtail (Cordis, Miami Lakes, FL) was placed at the site of the aortic valve. This assisted as a marker for site positioning the transeptal puncture, calibration during LAA contrast injection and monitoring of aortic pressures during the procedure. Calibration was performed using the 10 mm marker distance on a straight segment of the marker pigtail close to the LAA. Following attainment of right or left femoral venous access, transeptal puncture with a standard Mullins Sheath (Medtronic, Minneapolis, MN) and Brockenbrough needle (8F, Medtronic, Minneapolis, MN) was performed under TEE guidance with the site of puncture being aimed at the inferior portion of the membranous fossa ovalis, mid to posteriorly located. Use of alternating bicaval and short axis TEE views allowed for precise positioning. This puncture site facilitates subsequent orientation of the delivery system in the appendage and adequate space to maneuver the guide within the left atrium. In the three of first 10 patients, access to the LA was through a PFO. Subsequently, access to the LA was always through a transeptal puncture to improve catheter manipulation and orientation in the LA and LAA.
3921
A 5F multipurpose catheter with sideholes (Cordis, Miami Lakes, FL) was advanced though the Mullins sheath and using a combination of the angulations of the sheath and catheter under TEE and fluoroscopic guidance, maneuvered into the LAA. Selective contrast injections were performed in four standard views to interrogate the LAA. These views are AP, RAO 30°, RAO 30°/Cranial 15° and RAO 30°/Caudal 15°. In selected patients with impaired renal function, this was restricted to a single (RAO 30°/Cranial 15°) or occasionally two views. Quantitative angiographic analysis (Siemens Artis software, Munich, Germany) was performed in the best 2–3 views that demonstrated the LAA neck. A single frame demonstrating maximal appendage diameters was selected. A line was drawn across the LAA ostium. A second 10 mm line originating from the middle of the ostium and running in the axis of the LAA neck was then drawn. The measure of the landing zone was a third line perpendicular to the LAA neck axis running from the device engagement site beneath the circumflex coronary artery to the opposing wall (Fig. 2). Correlation of fluoroscopic sizing with the TEE measures within 2 mm was required prior to device selection. If a discrepancy between modality measures was noted, the measures were repeated including recalibration. Target correlation was achieved in all cases. The Amplatzer 45° × 45° TorqVue™ 3-dimensional delivery sheath (TorqueVue, AGA, Minneapolis MN) was used in almost all cases (97/100 due to availability). Sheath caliber selection was based upon device size selection according to company recommendations. Exchange was performed with a stiff 0.035″ guidewire (002 Amplatzer wire AGA, Minneapolis MN) positioned alternatively in the LAA or left upper pulmonary vein (LUPV). Once the delivery sheath was positioned in the LAA at a depth of approximately 2 cm, the device was advanced forward to the sheath tip. Under TEE guidance the initial portion of the device was unsheathed allowing it to attain a ball shape, excluding the first 20 patients. In those patients the device was advanced from within the sheath. At approximately the midpoint of expansion (designated by radio opaque markers), the sheath was stabilized and the device advanced. This allowed the device to return to its intended shape (much like a car tire), without advancing forward and as such not risking perforation. The device lobe position was then assessed by echo and contrast fluoroscopy and if the position was satisfactory the device disk was unsheathed within the LA and when exposed allowed to retract to its position covering the LAA ostium (Fig. 3). Due to extreme anatomical variability, device lobe positioning was not always as anticipated. If the operator interpreted that the lobe position was off angle, too deep or too shallow or if the disk position was considered too deep or disruptive to other structures, the device was resheathed and redeployed in a more favorable position. With position assessed by echo and contrast fluoroscopy, five criteria (which have subsequently been included in the company IFU) were used to confirm stability and adequate position: 1. Adequate lobe compression such that it resembles a car tire (and not a hockey puck as when unrestrained) 2. Lobe orientation being perpendicular to the LAA neck axis 3. Adequate separation of the device lobe and disk (as compared to the unrestrained position) 4. Device disk concavity, representing tension from the device lobe transferred though the connecting waist and improved likelihood of disk occlusion of the LAA ostium 5. Two thirds of the device lobe was located distal to the circumflex artery as visualized by TEE. If it was clear that the device size was inappropriate the device was retrieved, sheath changed and an alternate device deployed. When the above five criteria were achieved by both TEE and fluoroscopic imaging the device was released by unscrewing the delivery cable. Initially a “tug test” was performed to confirm stability prior to release. This was abandoned after the first 15 patients and the device released once the five release criteria were confirmed. Repeat transthoracic echocardiography was performed the next day to confirm device position and exclude embolization. In the event that the device could not be clearly discerned by echocardiography, fluoroscopy was performed to exclude embolization.
Fig. 1. Echocardiographic measurement of left atrial appendage (LAA) for device sizing at device lobe landing zone (dashed line) at 1 cm depth (solid line) from the appendage ostium (dotted line).
3922
D. Meerkin et al. / International Journal of Cardiology 168 (2013) 3920–3925 2.4. Statistical analysis Events are expressed as absolute numbers or percentages. Continuous variables are expressed as mean ± standard deviation or ranges. The authors of this manuscript have certified that they comply with the Principles of Ethical Publishing in the International Journal of Cardiology.
3. Results 3.1. Patient demographics
Fig. 2. Fluoroscopic measurement of left atrial appendage (LAA) for device sizing at device lobe landing zone (dashed line) at 1 cm depth from the appendage ostium (dotted line). This image was taken at an angulation of RAO 30°, Cranial 15°. PVR = Pulmonary vein ridge.
2.2. Anticoagulation All patients were treated with a standard heparin bolus dose of 5000–10,000 units administered after transeptal puncture with additional boluses to maintain an ACT of N250 s. All patients were treated with a loading dose of 300 mg of aspirin and 300 mg of clopidogrel post-procedure followed by a lifelong recommendation of 100 mg aspirin daily with 75 mg of clopidogrel recommended to be continued for one month only. Major adverse events were defined as death, stroke, systemic embolism, device embolization, pericardial bleeding requiring an intervention (cardiac tamponade) or other major bleeding requiring invasive treatment or blood transfusion occurring during index hospital admission. All procedures were performed by or under the direct supervision of a single operator (DM) and were performed in the absence of any formal surgical back-up. 2.3. Patient population The procedure was offered as a clinical indicated solution to patients with paroxysmal, persistent or permanent atrial fibrillation, a CHADS2 score of ≥2 and a contraindication or inability to be treated with Vitamin K antagonists. Anonymous patient and procedural information were included in a clinical database. This study is the retrospective analysis of consecutive patients treated by a single operator included in the clinical database. Forty-two cases were performed in a single center (SZMC, Jerusalem, Israel) and 58 were performed as demonstration or training cases in 23 centers in Europe, Australia and India. In the training cases, in-hospital outcomes were confirmed via contact with the managing team either by direct phone contact or through the local device distributor. Patients with LAA thrombus or aneurysm were excluded. In the advent of an LAA thrombus as assessed by TEE, if possible, the patient was treated with low molecular weight heparin for 3–4 weeks and returned for repeat assessment and occlusion. All patients signed informed consent for the procedure. Patients included in a randomized clinical trial of the device were also excluded from the analysis.
Between June 2009 and March 2012, one hundred patients underwent catheterization for closure of left atrial appendage with the Amplatzer Cardiac Plug (AGA, Minneapolis, US). Patient demographics are shown in Table 1. There were 55 males. The mean age was 73 ± 9.95 years. There were 48 patients ≥75 and 83 ≥ 65 years of age. 27 patients were ≥80 years of age. 58 patients suffered from permanent while 16 had persistent and 26 paroxysmal atrial fibrillation. 53 patients had previously suffered an ischemic neurological episode. Diabetes mellitus was present in 25 and hypertension in 94 patients. The mean CHADS2 score was 3.21 ± 1.23. 3.2. Contraindications The range of contraindications for chronic anticoagulation was broad including both absolute and relative contraindications as shown in Table 2. The most frequent reason for referral was bleeding with and without anticoagulants and the most frequent site was from the gastrointestinal tract. Lack of compliance with either medication or INR measures was also not infrequent at 14%. Additional, unusual causes included bone pain, allergies, residence in a remote area without INR measurement capabilities. 3.3. Landing zone measurements and device deployment Measures of the landing zone by TEE and by angiographic techniques are shown in Table 3. The mean landing zone as assessed by TEE was 20.01 ± 3.21 mm, while the mean landing zone measured by fluoroscopy usually at RAO 30° and RAO 30/Cranial 15° was 20.8 ± 3.19 mm. The mean difference between the TEE and the fluoroscopic measurements were 0.8 ± 1.13 mm. Based upon these measurements the device to be deployed and the delivery sheath were selected. In all cases the landing zone measurements of the LAA were such that the device closure with the ACP™ could be attempted. Device deployment was successful in 100/100 attempted cases. The mean device size deployed was 24.34 ± 3.27 mm. The deployed device sizes are shown in Fig. 4. Interestingly, 4 device sizes (22, 24, 26 and 28 mm) comprised 80% of the deployed devices representing a size range of 6 mm. In 18 cases the initially selected device incorrectly sized and the device removed and replaced with a second device. In
Fig. 3. ACP device deployment and position in the left atrial appendage. ACP = Amplatzer Cardiac Plug, Lt. Cx = left circumflex, PV = pulmonary vein.
D. Meerkin et al. / International Journal of Cardiology 168 (2013) 3920–3925
all these cases the second device was successfully deployed. There were no instances when a third device was taken. In 14 of these 18 cases the device was upsized. This was by 2 mm in 12 cases and 4 mm in 2 cases. The device was downsized in only 4 cases, three by 2 mm and in another case by 6 mm. With increased experience this was less frequent with only 5 of these 18 cases being in the second half of the patient series. Radiation time, contrast volume and post-procedural hospital stay were available for the cohort of patients treated in the single center (SZMC) and are presented in Table 4.
Table 2 Contraindications to anticoagulation. Bleeding GI Intracranial Ocular Other sources (epistaxis/respiratory etc.) Compliance Falls Sundry
67 40 15 2 11 14 9 11
Values are presented as number of patients. GI = gastrointestinal.
3.4. Complications There were two significant in-hospital complications (Table 4). The first was a pericardial effusion with tamponade that occurred during device deployment in a shallow LAA in a 79 year-old woman a CHADS2 score of 3 and severe lower GI bleeding (patient number 20). This occurred as the device was being pushed out of the delivery sheath. The complication was noted immediately, the device deployed to exclude the leak from the left atrium and the effusion was drained urgently. The patient recuperated well and was discharged on hospitalization day 6 with no further complications. The second complication occurred in an 82 year-old woman with a CHADS2 score of 4 and inability to control INR. This woman additionally suffered from severely impaired left ventricular systolic function and severe mitral incompetence with mild renal failure. Following an uncomplicated procedure during which she received 80 cm3 of nonionic contrast material, extubation was performed as routine. One hour later she developed acute respiratory distress with pulmonary edema. She was reintubated and repeat TEE performed demonstrating the ACP well in position and worsening of the mitral incompetence probably due to additional volume load. Coronary angiography was performed to exclude compression of the circumflex artery and was unremarkable. The patient responded well to diuretics and was extubated 6 h later and discharged on hospital day 4. 4. Discussion Two devices have clearly demonstrated the benefits of transcatheter occlusion of the left atrial appendage, the PLAATO device in published registries [10–12] and the Watchman device in the Protect AF trial [13] and subsequently the CAP registry [15]. The PLAATO device, the
Table 1 Patient demographics. Age (years) Male gender CHADS2a score CHADS2 score (%) 0 1 2 3 4 5 6 Congestive heart failure Hypertension Age N75 years Diabetes mellitus Previous embolic neurological event Atrial fibrillation pattern Paroxysmal Persistent Permanent
3923
73 ± 9.95 55 3.21 ± 1.23 0 7 24 30 21 16 2 49 94 48 25 53 26 16 58
Values are presented as number of patients or mean ± SD. a Clinical prediction rule for estimating the risk of stroke in patients with non-rheumatic atrial fibrillation (AF).
first of its class, demonstrated a reduction of expected events in registries that included high risk patients that were ineligible for oral anticoagulants [10–12]. The principal limitation of the device use was its relative high incidence of periprocedural complications and MACCE. For commercial reasons this device did not progress beyond the initial studies arresting the follow up in the longer-term registry [10]. The Watchman device, in a carefully planned and executed randomized trial demonstrated the non-inferiority of the technique when compared with oral anticoagulants in eligible patients. As a result, the patient population was less sick and therefore at lower stroke risk than the populations studied with the PLAATO device. The limitation of the technique as was manifest in PROTECT AF study appeared to be the relatively high rate of periprocedural complications [13]. These complications included device embolization, periprocedural stroke and air embolism, however the most frequent complication was significant pericardial effusion/tamponade. When assessing any new therapy the risk benefit ratio is one of the cardinal relationships that must be determined. As this important trial was performed in 59 sites in the US and Europe, a very significant portion of these patients were treated within the learning curve of the center and as such were exposed to higher risks of intraprocedural complications. The recent publication of the Continued Access Registry result from the 26 highest volume centers included in the PROTECT AF trial demonstrated that with refinement of technique and increased experience, a significant reduction of procedural complications could be achieved with the Watchman device [15]. The initial European experience with the Amplatzer Cardiac Plug™ was published last year and demonstrated the feasibility of the LAA occlusion with this device [14]. The ACP acts mechanistically according to a different principal than its predecessors and in distinction to its name, does not attempt to plug the LAA. It functions by placing a retainer in the LAA neck and covering its ostium with a second disk shaped element much as a lid would to a box. The patient population in this early experience was high risk and predominantly anticoagulant ineligible. This population is at particularly high risk for embolic stroke, bleeding complications and intraprocedural complications. As the emphasis of that publication was on the technical aspects of the device, much of the patient baseline data was not reported. Additionally, the report included 137 patients treated at 10 sites, such that each center had limited experience, much as occurred in the PROTECT AF trial. The safety findings from this early experience reported to 24 h demonstrated an event rate of 7% including pericardial effusions requiring drainage (3.5%) and embolization (1.4%). This correlated, with that of the PROTECT AF, where a higher than desirable number of intraprocedural complications were reported, predominantly driven by pericardial effusions requiring drainage (4.8%, embolization 0.6%), resulting in a complication rate of 10.2%. As an extension to this initial experience we present our safety and feasibility results to further define the true risk of the procedure in experienced hands as well as the further definition of the high risk population that are candidates for this therapy in our clinical practice. Our population included patients at particularly high risk for atrial fibrillation associated embolic stroke with a mean CHADS2 score of 3.24 ± 1.20, with the vast majority of patients over the age of 65 and
3924
D. Meerkin et al. / International Journal of Cardiology 168 (2013) 3920–3925
Table 3 Procedural demographics. Flouro landing zone size (mm) TEE landing zone size (mm) Delta flouro-TEE ACP initial selection size (mm) ACP deployed size (mm) Guide size (F)
Table 4 Procedural outcomes. 20.8 20.01 0.8 24.24 24.36 11.89
± ± ± ± ± ±
3.19 3.21 1.13 3.23 3.27 1.5
Values are presented as mean ± SD. ACP = Amplatzer Cardiac Plug, TEE = transesophageal echo.
at least half of them post-previous stroke. Their frailty, although not measured in the CHADS2 score is more manifest by the causes of anticoagulant ineligibility, the leading cause being GI bleeding. In spite of an extremely high preprocedural risk for complications, our complication rates were low and the procedure was extremely well tolerated. In the Continued Access Registry of the Watchman device, improved technique and patient selection allowed for a 95% implant success rate. In our current report with the ACP™, there were no cases in which the device was found inappropriate for patient anatomy, being able to be deployed in all attempted cases. Exclusion of cases was not based upon LAA measures but rather, upon extremely poor patient condition and LAA thrombus. Theoretically however, extremely large (landing zone N 28.5 mm) and small (landing zone b 12.5 mm) measurements would have been excluded. A single case of LAA aneurysm (N6 cm) was excluded from the analysis. These encouraging results further justify the use of this procedure in this difficult population and the paucity of complications shifts the risk benefit interaction away from the status of a procedure with a high complication rate with associated morbidities. An alternative role for this procedure may also be as an interim measure for patients who temporarily require the discontinuation of oral anticoagulants due to bleeding or surgical interventions. As the safety profile improves the risk for the short-term benefit is reduced and it may present as an attractive alternative. In spite of the acceptance that the LAA occlusion is effective for embolic protection in atrial fibrillation as a class effect, long-term follow-up is required to confirm the level of efficacy of this device, with particular emphasis on device stability, thrombus formation, device leaks and stroke rates. These data will be necessary to allow for the further expansion of the indications of the device to patients who are candidates for long-term anticoagulants.
Successful deployment Deployment criteria Tire shape Separation Disk concavity Adequate orientation Lobe 2/3 behind Cxa (echo measurement) Residual flow (behind device lobe) TEEb color flow Doppler Fluoroscopic direct contrast injection Radiation time (min)c Contrast volume (cm3)c Procedural complications Death Stroke Myocardial infarction Air embolism Device embolism Pericardial effusion Post-procedural hospital stay (days)c Mean Median
100 99 100 100 100 100 0 4 15.4 ± 7.1 95.8 ± 42.0 0 0 0 0 0 1 2.4 ± 4.1 1
Values are presented as number of patients or mean ± SD. a Cx = circumflex artery. b TEE = transesophageal echo. c n = 42 patients treated in SZMC.
4.1. Limitations Occlusion of the left atrial appendage has inherent risks due to the required transeptal puncture and subsequent hardware manipulation within the appendage with potentially fragile appendage and extremely thin wall. A retrospective single operator study, although able to demonstrate the results achievable by an experienced operator, is limited in its implications to the broader penetration of the procedure. Our results however do include the learning curve, which seems to be principally manifest by errors in sizing and fewer occurrences of significant complications. This initial report furthermore only focuses on the acute results of device deployment and periprocedural complications. The true efficacy of the device must be measured by stroke prevention, residual leaks and device related thrombus in the longer term, in combination with these periprocedural results. However due to the presence of previous publications with the ACP™ and other devices describing relatively high early complication rates, we felt that reporting the results achieved with specific technical guidelines could potentially contribute to a significant reduction in complication rates achieved with the ACP™. Our long-term results specifically targeted at clinical and echocardiographic outcomes will be reported subsequently. 5. Conclusions In this single operator report, left atrial appendage occlusion using the double element ACP™ can be safely performed with excellent success rates. Using very specific deployment success, stability and release criteria, this device can achieve LAA occlusion with extremely low complication rates in an extremely frail oral anticoagulant ineligible population with multiple co-morbidities. The efficacy of this device needs to be assessed with longer-term clinical and echocardiographic follow-up. The low peri-procedural complication rate however, offers promise for the application of this device in the oral anticoagulant eligible population as well as the high risk, no-option population presented in this report. References
Fig. 4. ACP device size distribution range.
[1] Hart RG, Pearce LA, Aguilar MI. Meta-analysis: antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial fibrillation. Ann Intern Med 2007;146:857–67. [2] European Heart Rhythm Association, European Association for Cardio-Thoracic Surgery, Camm AJ, et al. Guidelines for the management of atrial fibrillation: the
D. Meerkin et al. / International Journal of Cardiology 168 (2013) 3920–3925
[3]
[4] [5] [6] [7] [8]
[9]
Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Eur Heart J 2010;31:2369–429. Fuster V, Rydén LE, Cannom DS, et al. ACCF/AHA/HRS focused updates incorporated into the ACC/AHA/ESC 2006 Guidelines for the management of patients with atrial fibrillation: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines developed in partnership with the European Society of Cardiology and in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. J Am Coll Cardiol 2011;57: e101–98. Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med 2009;361:1139–51. Patel MR, Mahaffey KW, Garg J, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med 2011;365:883–91. Granger CB, Alexander JH, McMurray JJ, et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2011;365:981–92. Bungard TJ, Ghali WA, Teo KK, McAlister FA, Tsuyuki RT. Why do patients with atrial fibrillation not receive warfarin? Arch Intern Med 2000;160:41–6. Nieuwlaat R, Capucci A, Camm AJ, et al. Atrial fibrillation management: a prospective survey in ESC member countries: the Euro Heart Survey on atrial fibrillation. Eur Heart J 2005;26:2422–34. Blackshear JL, Odell JA. Appendage obliteration to reduce stroke in cardiac surgical patients with atrial fibrillation. Ann Thorac Surg 1996;61:755–9.
3925
[10] Bayard YL, Omran H, Neuzil P, et al. PLAATO (Percutaneous Left Atrial Appendage Transcatheter Occlusion) for prevention of cardioembolic stroke in nonanticoagulation eligible atrial fibrillation patients: results from the European PLAATO study. EuroIntervention 2010;6:220–6. [11] Park JW, Leithäuser B, Gerk U, Vrsansky M, Jung F. Percutaneous left atrial appendage transcatheter occlusion (PLAATO) for stroke prevention in atrial fibrillation: 2-year outcomes. J Invasive Cardiol 2009;21:446–50. [12] Ostermayer SH, Reisman M, Kramer PH, et al. Percutaneous left atrial appendage transcatheter occlusion (PLAATO system) to prevent stroke in high-risk patients with non-rheumatic atrial fibrillation: results from the international multi-center feasibility trials. J Am Coll Cardiol 2005;46:9–14. [13] Holmes DR, Reddy VY, Turi ZG, et al. Percutaneous closure of the left atrial appendage versus warfarin therapy for prevention of stroke in patients with atrial fibrillation: a randomised non-inferiority trial. Lancet 2009;374:534–42. [14] 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;77:700–6. [15] Reddy VY, Holmes D, Doshi SK, Neuzil P, Kar S. Safety of percutaneous left atrial appendage closure: results from the Watchman Left Atrial Appendage System for Embolic Protection in Patients with AF (PROTECT AF) clinical trial and the Continued Access Registry. Circulation 2011;123:417–24.