Catheterization and Cardiovascular Interventions 84:1190–1196 (2014)
Amplatzer Left Atrial Appendage Occlusion Through a Patent Foramen Ovale Dezsoe Koermendy, MD, Fabian Nietlispach, MD, Samera Shakir, MD, Steffen Gloekler, MD, Peter Wenaweser, MD, Stephan Windecker, MD, Ahmed A. Khattab, MD, and Bernhard Meier,* MD Objectives: To assess feasibility and outcomes of left atrial appendage (LAA) closure when using a patent foramen ovale (PFO) for left atrial access. Background: Because of the fear of entering the left atrium too high, using a PFO for left atrial access during LAA occlusion (LAAO) is generally discouraged. We report our single-center experience using a concomitant PFO for LAAO, thereby avoiding transseptal puncture. Methods: LAAO was performed with local anesthesia and fluoroscopic guidance only (no echocardiography). The Amplatzer Cardiac Plug (ACP) was used in all patients. After LAAO, the PFO was closed at the same sitting, using an Amplatzer occluder through the ACP delivery sheath. Patients were discharged the same or the following day on dual antiplatelet therapy for 1–6 months, at which time a follow-up transesophageal echocardiogram (TEE) was performed. Results: In 49 (96%) of 51 patients (35 males, age 70.9 6 11.9 years), LAAO was successful using the PFO for left atrial access. In one patient, a long tunnel PFO precluded LAAO, which was performed via a more caudal transseptal puncture. In a second patient, a previously inserted ASD occluder precluded LAAO, which was abandoned because of pericardial bleeding. PFO closure was successful in all patients. Follow-up TEE was performed in 43 patients 138 6 34 days after the procedure. It showed proper sitting of both devices in all patients. Conclusions: Using a PFO for LAAO had a high success rate and could be the default access in all patients with a PFO, potentially reducing procedural complications arising from transseptal puncture. VC 2014 Wiley Periodicals, Inc. Key words: left atrial appendage closure; patent foramen ovale/atrial septal defect; transseptal catheterization
INTRODUCTION
Percutaneous left atrial appendage occlusion (LAAO) evolved as an alternative treatment to oral anticoagulation (OAC) in patients with atrial fibrillation [1–7]. Early procedural complications during LAAO are subsequently outweighed by less bleeding complications with an at least similar reduction in cardioembolic events [4,8,9]. Four-year data from the randomized PROTECT-AF trial for the first time not only showed noninferiority but superiority of LAAO over medical therapy with a 32% relative risk reduction (RRR) (absolute risk reduction (ARR) 0.7%) and a significant 60% RRR (ARR 1.4%) in cardiovascular death of LAAO over medical therapy (D. Holmes, presented at EuroPCR 2013). However, given the flat learning curve of the procedure, the importance of further improving procedural safety remains. LAAO is considered a complex procedure with a rather flat [10] learning curve. Our frugal approach using local anesthesia and fluoroscopic guidance only C 2014 Wiley Periodicals, Inc. V
[8] has stood the test of time [2] and saves time and costs. To further improve procedural outcome, we investigated the use of the patent foramen ovale (PFO) Department of Cardiology, Bern University Hospital, Bern, Switzerland Dezsoe Koermendy and Fabian Nietlispach contributed equally to this work. Conflict of interest: Ahmed Khattab and Bernhard Meier received research grants and proctor fees from St. Jude Medical. Stephan Windecker received research grants from St. Jude Medical. *Correspondence to: Bernhard Meier, MD; Professor and Head of Cardiology, Swiss Cardiovascular Center Bern, Bern University Hospital, CH-3010 Bern, Switzerland. E-mail: [email protected] Received 24 June 2013; Revision accepted 30 December 2013 DOI: 10.1002/ccd.25354 Published online 8 January 2014 in Wiley Online Library (wileyonlinelibrary.com)
LAAO Through a PFO
1191
for LAAO. Not to use the PFO for left atrial access is a generally accepted doctrine based on the concern that the PFO is located too high, i.e., too cranially in the interatrial septum, thereby rendering proper intubation of the left atrial appendage (LAA) with the delivery sheath difficult. On the other hand, the PFO facilitates left atrial access and avoids potential complications of transseptal puncture [11].
MATERIALS AND METHODS Patients All patients undergoing LAAO with the Amplatzer Cardiac Plug (ACP, St. Jude Medical, Plymouth, MN) at our institution who were known to have a PFO or an atrial septal defect (ASD) were included in the analysis. Indications for LAAO comprised of a history of previous bleeding, patients at high-risk of bleeding, unstable INR, high risk or history of falls, or patient preference. All patients gave written informed consent before the procedure according to the stipulations and approval of the institutional Internal Review Board.
Procedure A transesophageal echocardiogram (TEE) was performed before the procedure, except in patients with ad hoc LAAO added to a cardiac catheterization for another reason. All procedures were performed under local anesthesia and with fluoroscopic guidance only (no intraprocedural echocardiography). At the beginning of the procedure, 5,000 units of heparin were given. The PFO or ASD was looked for and crossed either with the standard J-tip wire or by using a right Judkins or Multipurpose catheter. In case ad-hoc LAAO was performed, the LAAO was depicted using contrast injection through the Judkins or Multipurpose catheter to exclude LA thrombus. The catheter was then replaced by the 13 French Amplatzer TorqVue delivery sheath (St. Jude Medical, Plymouth, MN) over a 0.035-inch Backup wire (Boston Scientific, Natick, MA). Contrast injections through this sheath were used to depict the anatomy and the size of the LAA. In different projections (typically right anterior oblique (RAO) caudal and RAO cranial), the LAA was depicted and the outer diameter of the sheath (5.1 mm) was used as a reference diameter for device sizing, aiming to oversize the device by at least 20%. The device was deployed and a stable position confirmed angiographically and by a tug test. Then, the
Fig. 1. Final result after LAA and PFO closure. Left atrial appendage (LAA) occlusion through a patent foramen ovale (PFO) with PFO closure at the end through a single 13-French sheath. LA, left atrium; RA, right atrium.
device was released and the sheath slightly retracted, but kept in the left atrium. A properly sized Amplatzer PFO or ASD occluder (St. Jude Medical, Plymouth, MN) was introduced and deployed to the PFO or ASD using the ACP sheath and delivery cable. Sizing of the PFO occluder depended on the previously performed TEE, or on angiographic imaging in case of ad-hoc LAAO. ASDs were balloon sized. After confirming a safe position of septal occluder by contrast injections and a wiggle test, it was released (Fig. 1). Patients were discharged the day of the procedure or the following day. Before discharge, correct device positions were confirmed by a transthoracic echocardiography. Antithrombotic regimen consisted of acetylsalicylic acid (100 mg) and clopidogrel (75 mg) for 1–6 months after which time a follow-up TEE was performed to confirm proper seating of the devices and to look for thrombi or residual leaks. At the same time, a neurology consultation was performed to document potential new cerebrovascular events. ACP and Amplatzer PFO and Septal Occluders
All three devices are made of a double part nitinol mesh and are filled by polyester and connected by a waist. The ACP for LAAO consists of a lobe with
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
1192
Koermendy et al.
TABLE I. Baseline Characteristics
TABLE II.
Procedure Characteristics and Follow-up
No. (%) Age mean 6 SD, years Gender Male Female Type of arrhythmia Paroxysmal atrial fibrillation/flutter Persistent atrial fibrillation/flutter Atrial flutter Indication to LAA closure Previous bleeding High-risk of bleeding Labile INR Risk/history of falls Patient preference CHA2DS2-VASc score, mean 6 SD Congestive heart failure Hypertension Age 75 years Diabetes Stroke Vascular disease Age 65–75 years Female sex HAS-Bled score, mean 6 SD Hypertension >160 mm Hg Abnormal renal function Abnormal liver function Stroke Bleeding Labile INR Elderly Drug/alcohol abuse Cardiovascular risk factors Hypertension Diabetes Family history Dyslipidemia Obesity Smoking
71 6 12 35 (69%) 16 (31%) 28 (55%) 19 (37%) 4 (8%) 11 (22%) 18 (35%) 5 (10%) 2 (4%) 15 (29%) 2.5 6 1.4 10 (20%) 39 (77%) 24 (47%) 17 (33%) 20 (39%) 43 (84%) 14 (28%) 16 (31%) 2.4 6 1.3 2 (4%) 3 (6%) 4 (8%) 20 (39%) 20 (39%) 12 (24%) 38 (75%) 31 (61%) 10 17 8 25 13 16
(20%) (33%) (16%) (49%) (26%) (31%)
INR, international normalized ratio; LAA, left atrial appendage; SD, standard deviation.
No. (%) LAA þ PFO occlusion LAA þ ASD occlusion Occlusion during sinus rhythm Concurrent procedures Coronary angiography Percutaneous coronary intervention Transcatheter aortic valve implantation Ablation for atrial fibrillation ACP devices More than one device attempted 16 mm 18 mm 20 mm 22 mm 24 mm 26 mm 28 mm 30 mm PFO device Amplatzer PFO occluder 25 mm Amplatzer PFO occluder 18 mm Amplatzer cribriform septal occluder 18 Amplatzer cribriform septal occluder 25 Amplatzer cribriform septal occluder 30 ASD devices Amplatzer septal occluder 26 mm Amplatzer septal occluder 24 mm Amplatzer septal occluder 10 mm Amplatzer cribriform septal occluder 30 Same day discharge Clopidogrel >1 month Acetylsalicylic acid >6 months Discharge on coumadin Follow-up Clinical Death Major bleeding TEE Thrombus on ACP device Residual leak ACP Residual leak PFO
46 (90%) 5 (10%) 26 (52%)
mm mm mm
mm
38 17 4 3
(75%) (33%) (8%) (6%)
1 4 4 5 6 12 4 10 7
(2%) (8%) (8%) (10%) (12%) (24%) (8%) (20%) (14%)
34 2 1 7 2
(67%) (4%) (2%) (14%) (4%)
1 1 1 2 10 24 33 2
(2%) (2%) (2%) (4%) (20%) (47%) (65%) (4%)
51 5 3 35 3 1 5
(100%) (10%) (6%) (69%) (6%) (2%) (10%)
ACP, Amplazer cardiac plug; ASD, atrial septal defect; PFO, patent foramen ovale; TEE, transesophageal echocardiography.
anchoring hooks and a sealing disc. It is available in lobe sizes from 16 to 30 mm, requiring a 9- to 13French sheath. The dedicated TorqVue delivery sheath has two 45 out of plane bends designed for optimal engagement of the LAA. Because we exclusively relied on ad-hoc fluoroscopic device sizing, the 13-French sheath was used as default sheath in all patients, irrespective of the device implanted. For PFO closure, the Amplatzer PFO occluder or the cribriform ASD occluder was used. The Amplatzer PFO occluder comes in sizes 18–35 mm for the right atrial disc, with a smaller left atrial disc (except for the 18 mm or the Cribriform occluders). For ASD closure, a device was chosen as to exceed balloon sizing by at least 20%.
Patient and Procedural Characteristics
From January 2009 to 2013, 198 LAAOs were performed at our institution using the ACP in patients diagnosed with atrial fibrillation or flutter (Tables I and II). A PFO (46 patients) or ASD (five patients) was looked for in all and found in 51 patients (26%). They were used for the LAAO attempt in all of them. Cardiovascular comorbidities included heart failure in 10 (20%), arterial hypertension in 39 (77%), diabetes mellitus in 17 (33%), and a history of prior stroke or transient ischemic attack (TIA) in 20 (39%).
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
LAAO Through a PFO
1193
Fig. 2. Patient with previous ASD closure. Left: The LAA profile can be demonstrated with a Judkins right four catheter. Right: The ASD occluder deviates the 13-French delivery sheath to the roof of the left atrium, thereby precluding proper LAA engagement.
Endpoints and Statistics
Procedural success was defined as properly seated devices in the LAA and the PFO or ASD at the time of postprocedural TTE. Residual leaks were classified according to a flow jet >3 mm at the time of postprocedural TEE. Long-term follow-up was performed by telephone interview and review of patient charts in case of an event. Data are presented as mean and standard deviation (SD). Given the descriptive nature of this article, no between-groups comparisons were performed.
PFO was closed and pericardiocentesis was performed. No further attempt at LAAO was made (Fig. 2). PFO closure was successful in all patients. Procedural complications included air embolism with transient neurologic symptoms in one patient (2%), which had completely resolved at termination of the procedure. A hemodynamically irrelevant pericardial effusion was diagnosed the day after the procedure in two patients (4%) and was treated conservatively. A relevant pericardial effusion occurred in one patient (2%) during LAAO as described above. In a second patient, the LAA was occluded successfully, but the 13-French sheath was inadvertently pulled back into the right atrium before PFO closure. Recrossing the
RESULTS Procedural Outcome Procedural success was 98% (50 patients). In two patients (4%), the LAA was not successfully occluded using the PFO or ASD for left atrial access. In one of them, crossing the long tunnel PFO deviated the delivery sheath to the roof of the left atrium, precluding proper engagement of the LAA. PFO closure was performed during sheath retraction, and the LAA was occluded via a more caudal transseptal puncture. In the other patient, LAA occlusion was abandoned as per patient preference. The large retroverted LAA was difficult to reach because of a previously placed ASD occluder caudal to the PFO. This device deviated the sheath upward. Pericardial bleeding occurred after several unsuccessful attempts with two ACP sizes to stabilize a device in the LAA. The
TABLE III. Comparison Between LAAO via PFO/ASD and Transseptal Puncture
Procedural success Major procedural complications Device embolization Pericardial tamponade Major bleeding Stroke Transient ischemic attack Follow-up TEE available Leakage
via PFO/ASD
Transseptal puncture
P
50/51 (98%) 5/51 (10%)
143/147 (97%) 10/147 (7%)
0.83 0.340.72
0 (4%) (4%) (2%) (2%) (84%) (2%)
3/147 (2%) 4/147 (3%) 1/147 (1%) 1/147 (1%) 1/147 (1%) 105/147 (71%) 5/105 (5%)
0.96 0.33 0.58 0.97
2/51 2/51 0/51 1/51 43/51 1/43
0.07 0.96
TEE, transesophageal echocardiography. Dezsoe Koermendy and Fabian Nietlispach shared first authorship/equal contribution
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
1194
Koermendy et al.
PFO with the delivery sheath resulted in atrial wall injury and cardiac tamponade. Both patients with procedural tamponade (4%) were successfully treated with pericardiocentesis and discharged the following day. In comparison, of the 147 contemporarily treated patients with absence of congenital intraatrial shunt, a total of five hemodynamically relevant pericardial effusions occurred together with one puncture of the aortic root during transseptal puncture (6/147, 4.1%). In the latter, the aortic puncture was sealed with an ASD 5 mm occluder device. Overall procedural time for LAAO was if anything shorter when using an intra-atrial shunt for LA access, as compared with the group where a transseptal puncture was performed (80.6 min 6 22.3 vs. 84.5 min6 20.1, P ¼ 0.64).
Follow-Up
Clinical follow-up was complete in all patients at an average of 14.5 (69.7) months. During follow-up, five patients (two males, three females, age 79 6 10 years) died 130 6 61 days after LAA closure. Causes of death were noncardiac in all patients (pneumonia in 2, bleeding in 2, and lethal accident in 1 patient). Death due to pneumonia occurred 137 and 293 days after LAA closure. Death due to bleeding complications occurred 6 and 323 days after the procedure, respectively. The former patient died due to the gastrointestinal bleeding that was the reason for the hospitalization during which LAAO was performed. The latter died of an intracerebral bleed while on dual antiplatelet therapy (for percutaneous coronary intervention and transcatheter aortic valve implantation). The lethal accident happened 81 days after LAAO. In the remaining patients neither cerebrovascular, nor peripheral vascular embolic events occurred during a mean follow-up of 15 6 10 months. One patient experienced an intracerebral bleed while on OAC for a previously diagnosed thrombus on the ACP device. A follow-up TEE was available from 43 patients (84%) performed at an average of 138 6 34 days after the procedure. It showed proper position of both devices in all patients. There were three patients diagnosed with an LAA device thrombus during follow-up TEE. They were all put on a 3-month course of coumadin, after which another TEE was scheduled. One patient refused a repeat TEE and remained on OAC, while another patient experienced an intracerebral bleed while on coumadin (see above). In the third patient, the thrombus had resolved. There was one patient (2%) with a residual leak into the LAA. In comparison, there were five leaks in 147 patients of the cohort with trans-
septal puncture (3%). A residual shunt through the PFO was found in five patients (10%). DISCUSSION
The main findings of our study are (1) a high success rate of LAAO using a PFO or ASD for left atrial access, (2) a low rate of residual peri-device shunts into the LAA after LAAO using the ACP, and (3) feasibility of PFO or ASD closure using the ACP delivery system. LAA closure was successfully performed in >95% patients, avoiding transseptal puncture by using a PFO or ASD for left atrial access. This indicates that concerns using a PFO or ASD for LAAO may be unnecessary. Avoiding a transseptal puncture can obviate related complications and despite the fact that we performed an additional procedure in these patients (shunt-closure), overall procedural time was shorter when using such an approach. Complications associated to transseptal puncture in arrhythmia ablation interventions are low (80% after 18 months regarding all kind of interventions with transseptal puncture up to catheter size of 22 French [12]. Although the pathogenic impact of the iatrogenic septal defects might be low compared to the PFO, which is
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
LAAO Through a PFO
located exactly in the lower body inflow path, they could potentially mediate paradoxical embolism, pulmonary hypertension, or right ventricular dysfunction. There was no significant difference in the amount of residual leaks in the group where we used an intraatrial shunt for LAA access (2%) as compared with the group with transseptal puncture (5%) (Table III). Our overall rate of residual leaks is 4% (6/148 patients with TEE) less than that reported using the Watchman device even at a lower definition threshold (3 vs. 5 mm) [13]. When compared to all our patients who underwent LAAO using a transseptal puncture [8], major procedural complications were not more common in the heredescribed cohort. Pericardial bleeding was not reduced but not increased either by using an atrial defect for LAAO. However, long tunnel PFOs may represent anatomic variations rendering LAA engagement difficult, and transseptal puncture should be performed if initial attempts to deploy the ACP are unsuccessful. In patients where PFO closure was performed beforehand, the PFO occluder serves as an anatomic marker to guide transseptal puncture [14]. It goes without saying that closure of the LAA via a PFO or ASD is feasible irrespective of whether the interatrial septum defect is closed or not at the same sitting. Closure of a PFO or ASD is established as an alternative to anticoagulants in secondary prevention of cryptogenic stroke [15–17] with a >60% RRR [15,16] and a survival benefit [17]. We believed that because the delivery system is already in place, closure of the PFO at the same sitting makes sense. In our population, closure of the PFO or ASD was generally performed for primary prevention. The occlusion of the PFO or ASD in this setting is debatable and needs further studies, but addressing this subject was not an aim of this study. The patient with a relevant pericardial effusion after re-crossing the PFO with the 13-French sheath had suffered a previous TIA, thus his PFO was closed for secondary prevention. LIMITATIONS
Our study is descriptive and not a randomized trial. Given the study size, we are able to show feasibility of LAAO via a PFO but we cannot prove safety of such an approach. In our study, LAAO with the ACP resulted in a low number of peri-device leaks; however, the study size only allows generating the hypothesis that peri-device leaks are low when using the ACP and that this may be due to the pacifier principle provided by the proximal disk [5,7]. A larger study on peri-device leaks when using different devices is needed.
1195
CONCLUSIONS
There is a high success rate of LAA occlusion when using a PFO or ASD for left atrial access. Therefore, a PFO or ASD could be used as default for left atrial access. REFERENCES 1. Sievert H, Lesh MD, Trepels T, et al. Percutaneous left atrial appendage transcatheter occlusion to prevent stroke in high-risk patients with atrial fibrillation: Early clinical experience. Circulation 2002;105:1887–1889. 2. 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. 3. 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. 4. 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 noninferiority trial. Lancet 2009;374:534–542. 5. 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. 6. Lam YY, Yip GW, Yu CM, et al. Left atrial appendage closure with Amplatzer cardiac plug for stroke prevention in atrial fibrillation: Initial Asia-Pacific experience. Catheter Cardiovasc Interv 2012;79:794–800. 7. Nietlispach F, Gloekler S, Khattab A, et al. Percutaneous left atrial appendage closure. Eur J Geriatric Med 2012;3:308– 311. 8. 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. 9. Gangireddy SR, Halperin JL, Fuster V, et al. Percutaneous left atrial appendage closure for stroke prevention in patients with atrial fibrillation: An assessment of net clinical benefit. Eur Heart J 2012;33:2700–2708. 10. Reddy VY, Holmes D, Doshi SK, et al. 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;123:417–424. 11. Roelke M, Smith AJ, Palacios IF. The technique and safety of transseptal left heart catheterization: The Massachusetts general hospital experience with 1,279 procedures. Catheter Cardiovasc Diag 1994;32:332–339. 12. McGinty PM, Smith TW, Rogers JH. Transseptal left heart catheterization and the incidence of persistent iatrogenic atrial septal defects. J Interv Cardiol 2011;24:254–263. 13. Viles-Gonzalez JF, Kar S, Douglas P, et al. The clinical impact of incomplete left atrial appendage closure with the Watchman device in patients with atrial fibrillation: A PROTECT AF (percutaneous closure of the left atrial appendage versus Warfarin therapy for prevention of stroke in patients with atrial fibrillation) substudy. J Am Coll Cardiol 2012;59: 923–929.
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
1196
Koermendy et al.
14. Zaker-Shahrak R, Fuhrer J, Meier B. Transseptal puncture for catheter ablation of atrial fibrillation after device closure of patent foramen ovale. Catheter Cardiovasc Interv 2008;71: 551–552. 15. Meier B, Kalesan B, Mattle HP, et al. Percutaneous closure of patent foramen ovale in cryptogenic embolism. New Engl J Med 2013;368:1083–1091.
16. Carroll JD, Saver JL, Thaler DE, et al. Closure of patent foramen ovale versus medical therapy after cryptogenic stroke. New Engl J Med 2013;368:1092–1100. 17. Wahl A, J€uni P, Mono ML, et al. Long-term propensity scorematched comparison of percutaneous closure of patent foramen ovale with medical treatment after paradoxical embolism. Circulation 2012;125:803–812.
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
Amplatzer Left Atrial Appendage Occlusion Through a Patent Foramen Ovale Dezsoe Koermendy, MD, Fabian Nietlispach, MD, Samera Shakir, MD, Steffen Gloekler, MD, Peter Wenaweser, MD, Stephan Windecker, MD, Ahmed A. Khattab, MD, and Bernhard Meier,* MD Objectives: To assess feasibility and outcomes of left atrial appendage (LAA) closure when using a patent foramen ovale (PFO) for left atrial access. Background: Because of the fear of entering the left atrium too high, using a PFO for left atrial access during LAA occlusion (LAAO) is generally discouraged. We report our single-center experience using a concomitant PFO for LAAO, thereby avoiding transseptal puncture. Methods: LAAO was performed with local anesthesia and fluoroscopic guidance only (no echocardiography). The Amplatzer Cardiac Plug (ACP) was used in all patients. After LAAO, the PFO was closed at the same sitting, using an Amplatzer occluder through the ACP delivery sheath. Patients were discharged the same or the following day on dual antiplatelet therapy for 1–6 months, at which time a follow-up transesophageal echocardiogram (TEE) was performed. Results: In 49 (96%) of 51 patients (35 males, age 70.9 6 11.9 years), LAAO was successful using the PFO for left atrial access. In one patient, a long tunnel PFO precluded LAAO, which was performed via a more caudal transseptal puncture. In a second patient, a previously inserted ASD occluder precluded LAAO, which was abandoned because of pericardial bleeding. PFO closure was successful in all patients. Follow-up TEE was performed in 43 patients 138 6 34 days after the procedure. It showed proper sitting of both devices in all patients. Conclusions: Using a PFO for LAAO had a high success rate and could be the default access in all patients with a PFO, potentially reducing procedural complications arising from transseptal puncture. VC 2014 Wiley Periodicals, Inc. Key words: left atrial appendage closure; patent foramen ovale/atrial septal defect; transseptal catheterization
INTRODUCTION
Percutaneous left atrial appendage occlusion (LAAO) evolved as an alternative treatment to oral anticoagulation (OAC) in patients with atrial fibrillation [1–7]. Early procedural complications during LAAO are subsequently outweighed by less bleeding complications with an at least similar reduction in cardioembolic events [4,8,9]. Four-year data from the randomized PROTECT-AF trial for the first time not only showed noninferiority but superiority of LAAO over medical therapy with a 32% relative risk reduction (RRR) (absolute risk reduction (ARR) 0.7%) and a significant 60% RRR (ARR 1.4%) in cardiovascular death of LAAO over medical therapy (D. Holmes, presented at EuroPCR 2013). However, given the flat learning curve of the procedure, the importance of further improving procedural safety remains. LAAO is considered a complex procedure with a rather flat [10] learning curve. Our frugal approach using local anesthesia and fluoroscopic guidance only C 2014 Wiley Periodicals, Inc. V
[8] has stood the test of time [2] and saves time and costs. To further improve procedural outcome, we investigated the use of the patent foramen ovale (PFO) Department of Cardiology, Bern University Hospital, Bern, Switzerland Dezsoe Koermendy and Fabian Nietlispach contributed equally to this work. Conflict of interest: Ahmed Khattab and Bernhard Meier received research grants and proctor fees from St. Jude Medical. Stephan Windecker received research grants from St. Jude Medical. *Correspondence to: Bernhard Meier, MD; Professor and Head of Cardiology, Swiss Cardiovascular Center Bern, Bern University Hospital, CH-3010 Bern, Switzerland. E-mail: [email protected] Received 24 June 2013; Revision accepted 30 December 2013 DOI: 10.1002/ccd.25354 Published online 8 January 2014 in Wiley Online Library (wileyonlinelibrary.com)
LAAO Through a PFO
1191
for LAAO. Not to use the PFO for left atrial access is a generally accepted doctrine based on the concern that the PFO is located too high, i.e., too cranially in the interatrial septum, thereby rendering proper intubation of the left atrial appendage (LAA) with the delivery sheath difficult. On the other hand, the PFO facilitates left atrial access and avoids potential complications of transseptal puncture [11].
MATERIALS AND METHODS Patients All patients undergoing LAAO with the Amplatzer Cardiac Plug (ACP, St. Jude Medical, Plymouth, MN) at our institution who were known to have a PFO or an atrial septal defect (ASD) were included in the analysis. Indications for LAAO comprised of a history of previous bleeding, patients at high-risk of bleeding, unstable INR, high risk or history of falls, or patient preference. All patients gave written informed consent before the procedure according to the stipulations and approval of the institutional Internal Review Board.
Procedure A transesophageal echocardiogram (TEE) was performed before the procedure, except in patients with ad hoc LAAO added to a cardiac catheterization for another reason. All procedures were performed under local anesthesia and with fluoroscopic guidance only (no intraprocedural echocardiography). At the beginning of the procedure, 5,000 units of heparin were given. The PFO or ASD was looked for and crossed either with the standard J-tip wire or by using a right Judkins or Multipurpose catheter. In case ad-hoc LAAO was performed, the LAAO was depicted using contrast injection through the Judkins or Multipurpose catheter to exclude LA thrombus. The catheter was then replaced by the 13 French Amplatzer TorqVue delivery sheath (St. Jude Medical, Plymouth, MN) over a 0.035-inch Backup wire (Boston Scientific, Natick, MA). Contrast injections through this sheath were used to depict the anatomy and the size of the LAA. In different projections (typically right anterior oblique (RAO) caudal and RAO cranial), the LAA was depicted and the outer diameter of the sheath (5.1 mm) was used as a reference diameter for device sizing, aiming to oversize the device by at least 20%. The device was deployed and a stable position confirmed angiographically and by a tug test. Then, the
Fig. 1. Final result after LAA and PFO closure. Left atrial appendage (LAA) occlusion through a patent foramen ovale (PFO) with PFO closure at the end through a single 13-French sheath. LA, left atrium; RA, right atrium.
device was released and the sheath slightly retracted, but kept in the left atrium. A properly sized Amplatzer PFO or ASD occluder (St. Jude Medical, Plymouth, MN) was introduced and deployed to the PFO or ASD using the ACP sheath and delivery cable. Sizing of the PFO occluder depended on the previously performed TEE, or on angiographic imaging in case of ad-hoc LAAO. ASDs were balloon sized. After confirming a safe position of septal occluder by contrast injections and a wiggle test, it was released (Fig. 1). Patients were discharged the day of the procedure or the following day. Before discharge, correct device positions were confirmed by a transthoracic echocardiography. Antithrombotic regimen consisted of acetylsalicylic acid (100 mg) and clopidogrel (75 mg) for 1–6 months after which time a follow-up TEE was performed to confirm proper seating of the devices and to look for thrombi or residual leaks. At the same time, a neurology consultation was performed to document potential new cerebrovascular events. ACP and Amplatzer PFO and Septal Occluders
All three devices are made of a double part nitinol mesh and are filled by polyester and connected by a waist. The ACP for LAAO consists of a lobe with
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
1192
Koermendy et al.
TABLE I. Baseline Characteristics
TABLE II.
Procedure Characteristics and Follow-up
No. (%) Age mean 6 SD, years Gender Male Female Type of arrhythmia Paroxysmal atrial fibrillation/flutter Persistent atrial fibrillation/flutter Atrial flutter Indication to LAA closure Previous bleeding High-risk of bleeding Labile INR Risk/history of falls Patient preference CHA2DS2-VASc score, mean 6 SD Congestive heart failure Hypertension Age 75 years Diabetes Stroke Vascular disease Age 65–75 years Female sex HAS-Bled score, mean 6 SD Hypertension >160 mm Hg Abnormal renal function Abnormal liver function Stroke Bleeding Labile INR Elderly Drug/alcohol abuse Cardiovascular risk factors Hypertension Diabetes Family history Dyslipidemia Obesity Smoking
71 6 12 35 (69%) 16 (31%) 28 (55%) 19 (37%) 4 (8%) 11 (22%) 18 (35%) 5 (10%) 2 (4%) 15 (29%) 2.5 6 1.4 10 (20%) 39 (77%) 24 (47%) 17 (33%) 20 (39%) 43 (84%) 14 (28%) 16 (31%) 2.4 6 1.3 2 (4%) 3 (6%) 4 (8%) 20 (39%) 20 (39%) 12 (24%) 38 (75%) 31 (61%) 10 17 8 25 13 16
(20%) (33%) (16%) (49%) (26%) (31%)
INR, international normalized ratio; LAA, left atrial appendage; SD, standard deviation.
No. (%) LAA þ PFO occlusion LAA þ ASD occlusion Occlusion during sinus rhythm Concurrent procedures Coronary angiography Percutaneous coronary intervention Transcatheter aortic valve implantation Ablation for atrial fibrillation ACP devices More than one device attempted 16 mm 18 mm 20 mm 22 mm 24 mm 26 mm 28 mm 30 mm PFO device Amplatzer PFO occluder 25 mm Amplatzer PFO occluder 18 mm Amplatzer cribriform septal occluder 18 Amplatzer cribriform septal occluder 25 Amplatzer cribriform septal occluder 30 ASD devices Amplatzer septal occluder 26 mm Amplatzer septal occluder 24 mm Amplatzer septal occluder 10 mm Amplatzer cribriform septal occluder 30 Same day discharge Clopidogrel >1 month Acetylsalicylic acid >6 months Discharge on coumadin Follow-up Clinical Death Major bleeding TEE Thrombus on ACP device Residual leak ACP Residual leak PFO
46 (90%) 5 (10%) 26 (52%)
mm mm mm
mm
38 17 4 3
(75%) (33%) (8%) (6%)
1 4 4 5 6 12 4 10 7
(2%) (8%) (8%) (10%) (12%) (24%) (8%) (20%) (14%)
34 2 1 7 2
(67%) (4%) (2%) (14%) (4%)
1 1 1 2 10 24 33 2
(2%) (2%) (2%) (4%) (20%) (47%) (65%) (4%)
51 5 3 35 3 1 5
(100%) (10%) (6%) (69%) (6%) (2%) (10%)
ACP, Amplazer cardiac plug; ASD, atrial septal defect; PFO, patent foramen ovale; TEE, transesophageal echocardiography.
anchoring hooks and a sealing disc. It is available in lobe sizes from 16 to 30 mm, requiring a 9- to 13French sheath. The dedicated TorqVue delivery sheath has two 45 out of plane bends designed for optimal engagement of the LAA. Because we exclusively relied on ad-hoc fluoroscopic device sizing, the 13-French sheath was used as default sheath in all patients, irrespective of the device implanted. For PFO closure, the Amplatzer PFO occluder or the cribriform ASD occluder was used. The Amplatzer PFO occluder comes in sizes 18–35 mm for the right atrial disc, with a smaller left atrial disc (except for the 18 mm or the Cribriform occluders). For ASD closure, a device was chosen as to exceed balloon sizing by at least 20%.
Patient and Procedural Characteristics
From January 2009 to 2013, 198 LAAOs were performed at our institution using the ACP in patients diagnosed with atrial fibrillation or flutter (Tables I and II). A PFO (46 patients) or ASD (five patients) was looked for in all and found in 51 patients (26%). They were used for the LAAO attempt in all of them. Cardiovascular comorbidities included heart failure in 10 (20%), arterial hypertension in 39 (77%), diabetes mellitus in 17 (33%), and a history of prior stroke or transient ischemic attack (TIA) in 20 (39%).
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
LAAO Through a PFO
1193
Fig. 2. Patient with previous ASD closure. Left: The LAA profile can be demonstrated with a Judkins right four catheter. Right: The ASD occluder deviates the 13-French delivery sheath to the roof of the left atrium, thereby precluding proper LAA engagement.
Endpoints and Statistics
Procedural success was defined as properly seated devices in the LAA and the PFO or ASD at the time of postprocedural TTE. Residual leaks were classified according to a flow jet >3 mm at the time of postprocedural TEE. Long-term follow-up was performed by telephone interview and review of patient charts in case of an event. Data are presented as mean and standard deviation (SD). Given the descriptive nature of this article, no between-groups comparisons were performed.
PFO was closed and pericardiocentesis was performed. No further attempt at LAAO was made (Fig. 2). PFO closure was successful in all patients. Procedural complications included air embolism with transient neurologic symptoms in one patient (2%), which had completely resolved at termination of the procedure. A hemodynamically irrelevant pericardial effusion was diagnosed the day after the procedure in two patients (4%) and was treated conservatively. A relevant pericardial effusion occurred in one patient (2%) during LAAO as described above. In a second patient, the LAA was occluded successfully, but the 13-French sheath was inadvertently pulled back into the right atrium before PFO closure. Recrossing the
RESULTS Procedural Outcome Procedural success was 98% (50 patients). In two patients (4%), the LAA was not successfully occluded using the PFO or ASD for left atrial access. In one of them, crossing the long tunnel PFO deviated the delivery sheath to the roof of the left atrium, precluding proper engagement of the LAA. PFO closure was performed during sheath retraction, and the LAA was occluded via a more caudal transseptal puncture. In the other patient, LAA occlusion was abandoned as per patient preference. The large retroverted LAA was difficult to reach because of a previously placed ASD occluder caudal to the PFO. This device deviated the sheath upward. Pericardial bleeding occurred after several unsuccessful attempts with two ACP sizes to stabilize a device in the LAA. The
TABLE III. Comparison Between LAAO via PFO/ASD and Transseptal Puncture
Procedural success Major procedural complications Device embolization Pericardial tamponade Major bleeding Stroke Transient ischemic attack Follow-up TEE available Leakage
via PFO/ASD
Transseptal puncture
P
50/51 (98%) 5/51 (10%)
143/147 (97%) 10/147 (7%)
0.83 0.340.72
0 (4%) (4%) (2%) (2%) (84%) (2%)
3/147 (2%) 4/147 (3%) 1/147 (1%) 1/147 (1%) 1/147 (1%) 105/147 (71%) 5/105 (5%)
0.96 0.33 0.58 0.97
2/51 2/51 0/51 1/51 43/51 1/43
0.07 0.96
TEE, transesophageal echocardiography. Dezsoe Koermendy and Fabian Nietlispach shared first authorship/equal contribution
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
1194
Koermendy et al.
PFO with the delivery sheath resulted in atrial wall injury and cardiac tamponade. Both patients with procedural tamponade (4%) were successfully treated with pericardiocentesis and discharged the following day. In comparison, of the 147 contemporarily treated patients with absence of congenital intraatrial shunt, a total of five hemodynamically relevant pericardial effusions occurred together with one puncture of the aortic root during transseptal puncture (6/147, 4.1%). In the latter, the aortic puncture was sealed with an ASD 5 mm occluder device. Overall procedural time for LAAO was if anything shorter when using an intra-atrial shunt for LA access, as compared with the group where a transseptal puncture was performed (80.6 min 6 22.3 vs. 84.5 min6 20.1, P ¼ 0.64).
Follow-Up
Clinical follow-up was complete in all patients at an average of 14.5 (69.7) months. During follow-up, five patients (two males, three females, age 79 6 10 years) died 130 6 61 days after LAA closure. Causes of death were noncardiac in all patients (pneumonia in 2, bleeding in 2, and lethal accident in 1 patient). Death due to pneumonia occurred 137 and 293 days after LAA closure. Death due to bleeding complications occurred 6 and 323 days after the procedure, respectively. The former patient died due to the gastrointestinal bleeding that was the reason for the hospitalization during which LAAO was performed. The latter died of an intracerebral bleed while on dual antiplatelet therapy (for percutaneous coronary intervention and transcatheter aortic valve implantation). The lethal accident happened 81 days after LAAO. In the remaining patients neither cerebrovascular, nor peripheral vascular embolic events occurred during a mean follow-up of 15 6 10 months. One patient experienced an intracerebral bleed while on OAC for a previously diagnosed thrombus on the ACP device. A follow-up TEE was available from 43 patients (84%) performed at an average of 138 6 34 days after the procedure. It showed proper position of both devices in all patients. There were three patients diagnosed with an LAA device thrombus during follow-up TEE. They were all put on a 3-month course of coumadin, after which another TEE was scheduled. One patient refused a repeat TEE and remained on OAC, while another patient experienced an intracerebral bleed while on coumadin (see above). In the third patient, the thrombus had resolved. There was one patient (2%) with a residual leak into the LAA. In comparison, there were five leaks in 147 patients of the cohort with trans-
septal puncture (3%). A residual shunt through the PFO was found in five patients (10%). DISCUSSION
The main findings of our study are (1) a high success rate of LAAO using a PFO or ASD for left atrial access, (2) a low rate of residual peri-device shunts into the LAA after LAAO using the ACP, and (3) feasibility of PFO or ASD closure using the ACP delivery system. LAA closure was successfully performed in >95% patients, avoiding transseptal puncture by using a PFO or ASD for left atrial access. This indicates that concerns using a PFO or ASD for LAAO may be unnecessary. Avoiding a transseptal puncture can obviate related complications and despite the fact that we performed an additional procedure in these patients (shunt-closure), overall procedural time was shorter when using such an approach. Complications associated to transseptal puncture in arrhythmia ablation interventions are low (80% after 18 months regarding all kind of interventions with transseptal puncture up to catheter size of 22 French [12]. Although the pathogenic impact of the iatrogenic septal defects might be low compared to the PFO, which is
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
LAAO Through a PFO
located exactly in the lower body inflow path, they could potentially mediate paradoxical embolism, pulmonary hypertension, or right ventricular dysfunction. There was no significant difference in the amount of residual leaks in the group where we used an intraatrial shunt for LAA access (2%) as compared with the group with transseptal puncture (5%) (Table III). Our overall rate of residual leaks is 4% (6/148 patients with TEE) less than that reported using the Watchman device even at a lower definition threshold (3 vs. 5 mm) [13]. When compared to all our patients who underwent LAAO using a transseptal puncture [8], major procedural complications were not more common in the heredescribed cohort. Pericardial bleeding was not reduced but not increased either by using an atrial defect for LAAO. However, long tunnel PFOs may represent anatomic variations rendering LAA engagement difficult, and transseptal puncture should be performed if initial attempts to deploy the ACP are unsuccessful. In patients where PFO closure was performed beforehand, the PFO occluder serves as an anatomic marker to guide transseptal puncture [14]. It goes without saying that closure of the LAA via a PFO or ASD is feasible irrespective of whether the interatrial septum defect is closed or not at the same sitting. Closure of a PFO or ASD is established as an alternative to anticoagulants in secondary prevention of cryptogenic stroke [15–17] with a >60% RRR [15,16] and a survival benefit [17]. We believed that because the delivery system is already in place, closure of the PFO at the same sitting makes sense. In our population, closure of the PFO or ASD was generally performed for primary prevention. The occlusion of the PFO or ASD in this setting is debatable and needs further studies, but addressing this subject was not an aim of this study. The patient with a relevant pericardial effusion after re-crossing the PFO with the 13-French sheath had suffered a previous TIA, thus his PFO was closed for secondary prevention. LIMITATIONS
Our study is descriptive and not a randomized trial. Given the study size, we are able to show feasibility of LAAO via a PFO but we cannot prove safety of such an approach. In our study, LAAO with the ACP resulted in a low number of peri-device leaks; however, the study size only allows generating the hypothesis that peri-device leaks are low when using the ACP and that this may be due to the pacifier principle provided by the proximal disk [5,7]. A larger study on peri-device leaks when using different devices is needed.
1195
CONCLUSIONS
There is a high success rate of LAA occlusion when using a PFO or ASD for left atrial access. Therefore, a PFO or ASD could be used as default for left atrial access. REFERENCES 1. Sievert H, Lesh MD, Trepels T, et al. Percutaneous left atrial appendage transcatheter occlusion to prevent stroke in high-risk patients with atrial fibrillation: Early clinical experience. Circulation 2002;105:1887–1889. 2. 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. 3. 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. 4. 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 noninferiority trial. Lancet 2009;374:534–542. 5. 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. 6. Lam YY, Yip GW, Yu CM, et al. Left atrial appendage closure with Amplatzer cardiac plug for stroke prevention in atrial fibrillation: Initial Asia-Pacific experience. Catheter Cardiovasc Interv 2012;79:794–800. 7. Nietlispach F, Gloekler S, Khattab A, et al. Percutaneous left atrial appendage closure. Eur J Geriatric Med 2012;3:308– 311. 8. 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. 9. Gangireddy SR, Halperin JL, Fuster V, et al. Percutaneous left atrial appendage closure for stroke prevention in patients with atrial fibrillation: An assessment of net clinical benefit. Eur Heart J 2012;33:2700–2708. 10. Reddy VY, Holmes D, Doshi SK, et al. 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;123:417–424. 11. Roelke M, Smith AJ, Palacios IF. The technique and safety of transseptal left heart catheterization: The Massachusetts general hospital experience with 1,279 procedures. Catheter Cardiovasc Diag 1994;32:332–339. 12. McGinty PM, Smith TW, Rogers JH. Transseptal left heart catheterization and the incidence of persistent iatrogenic atrial septal defects. J Interv Cardiol 2011;24:254–263. 13. Viles-Gonzalez JF, Kar S, Douglas P, et al. The clinical impact of incomplete left atrial appendage closure with the Watchman device in patients with atrial fibrillation: A PROTECT AF (percutaneous closure of the left atrial appendage versus Warfarin therapy for prevention of stroke in patients with atrial fibrillation) substudy. J Am Coll Cardiol 2012;59: 923–929.
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
1196
Koermendy et al.
14. Zaker-Shahrak R, Fuhrer J, Meier B. Transseptal puncture for catheter ablation of atrial fibrillation after device closure of patent foramen ovale. Catheter Cardiovasc Interv 2008;71: 551–552. 15. Meier B, Kalesan B, Mattle HP, et al. Percutaneous closure of patent foramen ovale in cryptogenic embolism. New Engl J Med 2013;368:1083–1091.
16. Carroll JD, Saver JL, Thaler DE, et al. Closure of patent foramen ovale versus medical therapy after cryptogenic stroke. New Engl J Med 2013;368:1092–1100. 17. Wahl A, J€uni P, Mono ML, et al. Long-term propensity scorematched comparison of percutaneous closure of patent foramen ovale with medical treatment after paradoxical embolism. Circulation 2012;125:803–812.
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
