© 2014, Wiley Periodicals, Inc. DOI: 10.1111/joic.12138
STRUCTURAL HEART DISEASE Leaving No Hole Unclosed: Left Atrial Appendage Occlusion in Patients Having Closure of Patent Foramen Ovale or Atrial Septal Defect SAMEER GAFOOR, M.D.,1 JENNIFER FRANKE, M.D.,2 PATRICK BOEHM,1 SIMON LAM, M.D.,1 STEFAN BERTOG, M.D.,1,3 LAURA VASKELYTE, M.D.,1 ILONA HOFMANN, M.D.,1 and HORST SIEVERT, M.D., PH.D.1 From the 1Cardiovascular Center Frankfurt, Frankfurt, Germany; 2University of Heidelberg, Heidelberg, Germany; and 3Minneapolis Veterans Affairs Hospital, Minneapolis, Minnesota
Objectives: To report procedural outcome of sequential occlusion of the left atrial appendage (LAA) and an interatrial septal communication and discuss possible indications. Background: There are some patients who may have indications for both closure of patent foramen ovale (PFO) or atrial septal defect (ASD), as well as closure of the LAA. The optimal procedural strategy is not known. Methods: A retrospective review of LAA and PFO/ASD cases at our center was performed. Demographic, echocardiographic, and procedural data were recorded. Results: Closure of LAA and then PFO/ASD (Group I: 11 patients), closure of both in the same setting (Group II: 3 patients), and closure of the PFO/ASD and then the LAA (Group III: 3 patients) was performed in a total of 17 patients. Average age was 63.5 9.8 years. Most patients were hypertensive with prior cerebrovascular event in 52.9% of patients. Procedural success was 100%. Procedural adverse events were 2 episodes of tamponade (in Group I after first LAA procedure) treated with pericardiocentesis and 1 access‐site hematoma (Group II) treated conservatively. Conclusions: Sequential (staged or during the same procedure) closure of the LAA and interatrial communications can be performed safely in a carefully selected patient population. This is also the first known report of LAA occlusion in patients with prior septal closure devices. (J Interven Cardiol 2014;27:414–422)
Introduction Via communication between the venous and systemic circulation, both patent foramen ovale and atrial septal defects may allow paradoxical embolism to occur. Hence both have been implicated as the cause of strokes.1–3 Moreover, atrial septal defects are associated with a higher incidence of atrial fibrillation (AF). Given the risk of paradoxical and AF‐related stroke with interatrial communication, anticoagulation is often recommended in patients with additional AF. However, anticoagulation with warfarin is associated with both extracranial and intracranial hemorrhage4
Address for reprints: Dr. Horst Sievert, Cardiovascular Center Frankfurt, Seckbacher Landstrasse 65, 60389 Frankfurt am Main, Germany. Fax: þ49‐69‐46031343; e‐mail: [email protected]
414
and up to 40% of patients with AF have contraindications to anticoagulation.5 Newer anticoagulants have similar bleeding rates compared to warfarin with high rates of discontinuation.6 For these patients, closure of the left atrial appendage may have merit. In patients with both interatrial communication and AF, closure of the interatrial communication and left atrial appendage may be warranted, especially in those with high bleeding risk or contraindications to anticoagulation. When a cerebrovascular event occurs in this setting, it is difficult to say whether the cause is the interatrial communication or thrombus from AF. Little data exist evaluating the safety of closing both the LAA and an interatrial communication, with only 2 reports in the literature.7,8 The purpose of this study was to evaluate feasibility and safety of percutaneous closure of both the LAA and interatrial septal communication.
Journal of Interventional Cardiology
Vol. 27, No. 4, 2014
LEAVING NO HOLE UNCLOSED
Methods This was a retrospective review analysis of consecutive cases of both percutaneous interatrial septal communication repair (ASD/PFO) and left atrial appendage occlusion from 2002 to 2013. The patients were divided into 3 groups based on whether the patients had LAA occlusion prior, in the same setting, or after ASD/PFO closure. Key demographic, echocardiographic, and procedural data were collected for each patient. Follow‐up was performed using a combination of follow‐up visits as well as phone and mail survey assessment. Left Atrial Appendage Occlusion. Patients were selected for left atrial appendage closure based on a history of AF and inability to take anticoagulation or patient desire to be off of anticoagulation. Percutaneous left atrial appendage closure was performed with femoral venous access and transseptal puncture. Size of device was determined based on measurements of LAA from angiography and echocardiogram. Device was placed and sheath was removed. Patients had follow‐up echocardiography the next day and were discharged on a regimen of aspirin indefinitely and clopidogrel for 3– 6 months. Atrial Septal Defect/Patent Foramen Ovale Closure. Atrial septal defect was closed according to guidelines for congenital heart disease. Patent foramen ovale was closed for cryptogenic stroke. Via standard femoral venous access, balloon sizing was performed to evaluate the width of the atrial septal defect or the length of the patent foramen ovale tunnel. Concurrent transesophageal echocardiogram was used to evaluate the interatrial septum for other defects. The defect was percutaneously closed using an appropriate PFO or ASD occluder and then evaluated with transesophageal echocardiogram and angiography for residual shunt. Group 1: LAA Occlusion Followed by Interatrial Communication Closure in Separate Settings. For this group of patients, the procedure was performed similar to the techniques mentioned above. The puncture site for LAA occlusion may be through the existing interatrial communication (although some operators call for a new puncture site). Clinical setting dictated the follow‐up time period between the first intervention (LAA occlusion) and the second intervention (interatrial communication closure). During the interatrial communication closure, care was taken to
Vol. 27, No. 4, 2014
place the wire in the left upper pulmonary vein without disturbing the LAA occlusion device. This requires a good understanding of the left upper pulmonary vein and LAA anatomical relationship. Group II: LAA Occlusion and Interatrial Communication Closure in the Same Setting. The LAA occlusion was performed first, followed by interatrial communication closure. In this setting, the LAA occlusion was performed through the interatrial communication (Fig. 3). It is our protocol to wait 5– 10 minutes after LAA occlude placement before closing the PFO/ASD. This maintains access to the left atrium in case of early LAA occluder embolization. From our experience, this may occur even with good placement, adequate compression, and follow‐up angiography showing trace or minimal leak. One limitation of this method is that in the rare case of infection after the procedure, it is difficult to know whether the infected device is the LAA occlusion device, the interatrial septum closure device, both, or neither. Group III: PFO Closure Followed by LAA Occlusion in Separate Settings. After interatrial communication closure, the patient may develop atrial fibrillation and be a candidate for LAA closure. In this case, the procedure is more challenging as there is a device at the septum. It is our practice to attempt a puncture immediately inferior to the device. This is accomplished by observing the septum in the 30° and 90° view and puncturing inferior to the device and posterior to the aorta (Fig. 1). However, anterior puncture may also be used (Fig. 2), as long as the aorta is far away from the puncture site. Another option is to cross through the device, which is possible with some devices.9,10
Results Baseline demographic data are shown in Table 1. A total of 17 patients (mean age 63.5 years) were identified. Eleven patients first underwent closure of the LAA and then ASD/PFO (Group I), in 3 both procedures performed in the same setting (Group II), and in 3 patients ASD/PFO closure was performed first followed by LAA occlusion (Group III). In Group I, the time between LAA occlusion and ASD/PFO closure was 8.4 6.8 months, while for group III the time between ASD/PFO closure and LAA occlusion was 22.8 20.3 months. The majority of patients had paroxysmal AF, with a trend toward more permanent
Journal of Interventional Cardiology
415
GAFOOR, ET AL.
Figure 1. Placement of LAA occluder in a patient with prior ASD closure. The procedure began with a transseptal puncture inferior to the ASD occluder in the LAO view (A), followed by advancement of a pigtail catheter to take measurements of the LAA (B). This was then exchanged (C, D) for a large sheath and an LAA occluder was advanced (E) and deployed (F). At deployment one can see the relationship between the LAA occluder and ASD.
AF in Group III. One patient in Group I had cardioversion prior to the procedure. Pulmonary vein isolation or MAZE procedures were not performed in any of the patients. More than half of the patients had a history of thromboembolism, except for Group III in which only 1 patient had a history of stroke. CHADS2‐ Score was 2 across all Groups, and the mean CHA2DS2‐VASC‐Score was at least 3.9. Four of the 17 patients had previous bleeding complications on anticoagulation, and 1 had an allergic reaction. Therefore, just under half of patients were on anticoagulation, with the exception of Group III where 2 of 3 patients were on anticoagulation. Detailed echocardiographic data are shown in Table 2. The patients had normal left ventricular ejection fraction except for 1 patient. Approximately half of the patients had an atrial septal defect and the other half a patent foramen ovale. One patient in Group I had both. All patients with PFO had cryptogenic stroke except for 1 patient with PFO in Group I who had closure secondary to migraines. Another patient with PFO and cryptogenic
416
stroke in Group III was found to have subsequent AF that was not present at the time of first procedure. Table 3 shows procedural data for both procedures. Procedural success was 100%. For LAA occlusion, the Amplatzer device (ACP) was used in the vast majority of cases, followed by the PLAATO occluder. For patients with LAA occlusion followed by ASD/PFO closure, the ASD/PFO was used for access for the LAA occlusion. Group I had the most partial (3) and full (3) recaptures, of which 2 occurred in the same patient. For interatrial communication closure, the Amplatzer ASD/ PFO device was used in the vast majority (70.6%) of cases, followed by the Helex closure device (11.8%). Mean total fluoroscopy time was longer for the combined procedure (Group II) than individual procedures (Groups I and III). Procedural adverse events are shown in Table 4. These include pericardial effusion in 2 patients in Group I (treated successfully with pericardiocentesis) and a hematoma in 1 patient in Group II. Both adverse events in Group I happened in relation to the first procedure (LAA occlusion).
Journal of Interventional Cardiology
Vol. 27, No. 4, 2014
LEAVING NO HOLE UNCLOSED
Discussion
Figure 2. Transseptal procedure in patient with prior PFO device. Transesophageal echocardiogram at mid‐esophageal level showing transseptal procedure in patient with prior PFO device (left arrow). In this patient, the crossing point for the transseptal needle was in the anterior part of the septum (right arrow).
PFO (with cryptogenic stroke) and ASD are both risk factors for stroke,1,3 which is only increased by the presence of AF. One strategy in selected patients is closure of both the interatrial communication and left atrial appendage. To our knowledge, this is the first case series of such patients with both interatrial septal closure and left atrial appendage occlusion. There are both technical and indication‐based issues that would drive the order and necessity of both these procedures in the same patient. Technical Issues. The technique behind interatrial septal defect closure and LAA occlusion has been previously defined. This article highlights the ability to have a strategy where both procedures can be done in the same setting or in subsequent procedures. When closing the LAA first, the procedure is straight forward. When present, the PFO or ASD can be used for transseptal access. Some operators have called for a different and new transseptal puncture site for LAA
Figure 3. Placement of the LAA occluder and PFO closure device in the same setting. After crossing the septum through the PFO, balloon sizing was used to measure the orifice diameter (A). Next, a pigtail was placed in the LAA (B) and used to exchange for a delivery sheath to allow LAA occluder placement (C). Next, the same device sheath was used for PFO occluder delivery and deployment (D, E). The final position of the LAA occluder and PFO closure device is seen in the AP view (F).
Vol. 27, No. 4, 2014
Journal of Interventional Cardiology
417
GAFOOR, ET AL. Table 1. Baseline Characteristics
Age at first intervention (years) Time between interventions (months) Height (cm) Weight (kg) Hypertension Hyperlipidemia Diabetes Migraine Coronary artery disease Myocardial infarction Congestive heart failure NYHA class Peripheral vascular disease Deep vein thrombosis Atrial fibrillation Paroxysmal Persistent Permanent Cardioversion EP ablation MAZE procedure Previous TIA Previous CVA History of thromboembolism (TIA/CVA) CHADS2 CHADS2‐VASc Previous bleeding complication on anticoagulation (number of patients) GI bleeding Liver disease Other contraindication for vitamin K antagonist (VKA) Medications at time of first intervention Aspirin Plavix VKA
LAA Then ASD/PFO (11 Patients)
Concurrent (3 Patients)
ASD/PFO Then LAA (3 Patients)
Total (17 Patients)
63.4 8.8 8.44 6.8 169.1 10.4 78.8 17.7 90.9% 25% 9.1% 9.1% 27.3% 0% 27.3% 1.2 0.8 0% 0%
59 5.7 0 172 5.7 82.3 17.3 100% 0% 33% 0% 33% 0% 0% 0 0% 33.3%
63.3 16.7 22.8 20.3 181 8.5 93 22.6 100% 66.7% 66.7% 33.3% 66.7% 66.6% 66.7% 1.7 0.6 33.3% 33.3%
63.5 9.8 9.5 11.4 171 9.8 80.5 17.2 94.1% 35.3% 23.5% 11.7% 35.3% 11.7% 29.4% 1.1 0.8 6% 11.8%
63.6% 18.2% 18.2% 8.7% 0% 0% 36% 45% 55% 2.4 1.0 3.7 1.2 1 (hematoma, epistaxis) 0% 0% 0
66.6% 33.3% 0% 0% 0% 0% 50% 66.7% 66.7% 31 41 1 (retinal)
52.9% 23.5% 23.5% 0% 0% 0% 35.3% 47.1% 52.9% 2.6 0.9 3.9 1.1 4
0% 0% 0
0% 33.3% 66.6% 0% 0% 0% 0% 33.3% 33.3% 30 4.7 0.6 2 (gastrointestinal, epistaxis) 0% 0% 1 (allergic reaction)
18.2% 9.1% 36.4%
33.3% 0% 33.3%
0% 0% 66.6%
17.6% 5.9% 41.2%
0% 0%
CVA, cardiovascular accident; EP, electrophysiology; GI, gastrointestinal; NYHA, New York Heart Association; TIA, transient ischemic attack.
occlusion.11 In our experience, a new septal puncture is usually not necessary and the septum can be safely crossed through the PFO or ASD. Second, care must be taken to not dislodge the prior placed LAA device when inserting the PFO/ASD sheath. This requires a good understanding of the left upper pulmonary vein and left atrial appendage anatomy. When interatrial communication closure is followed by LAA occlusion, the procedure is more challenging. One may cross the septum around the device. This can be done in various areas; however, it is our practice to attempt a puncture immediately inferior to the device when possible
418
(Fig. 1). However, anterior puncture may also be used (Fig. 2). The other option includes crossing through the device, which from our experience is possible with some devices.9,10 To our knowledge, this is the first report of LAA occlusion in patients with prior septal closure devices. When both procedures are done in the same setting, it is easier to use the interatrial communication for advancement of the LAA sheath and device to be followed by ASD or PFO closure (Fig. 3). However, significant issues with this approach are also present. First, closure of the interatrial septum shortly after LAA occlusion would preclude quick access to the
Journal of Interventional Cardiology
Vol. 27, No. 4, 2014
LEAVING NO HOLE UNCLOSED Table 2. Echocardiographic Data
Ejection fraction (%) Valvular disease Mitral regurgitation (presence of) Interatrial communication ASD PFO Left atrial appendage multiple lobes (%) Maximal measurement at ostium (mm)
LAA Then ASD/PFO (n ¼ 11)
Concurrent (n ¼ 3)
ASD/PFO Then LAA (n ¼ 3)
Total (n ¼ 17)
67.9 10.7
64 3.6
49.3 9.8
63.1 11.7
63.6%
100%
66.6%
70.6%
54.4% 54.4%
33.3% 66.6%
66.6% 33.3%
52.9% 52.9%
45.5% 17.8 5.5
0% 22.1 1.4
33.3% 23.4 1.5
35.3% 19.8 4.9
One patient in this group had a hybrid defect of both ASD and PFO.
left atrium in case of device embolization. This could be avoided by placing the LAA occluder and then postponing PFO/ASD closure for 10–15 minutes to see if the device is prone to early embolization. Second, in the rare case of infection after the procedure, it is difficult to know whether the infected device is the LAA occlusion device, the interatrial septum closure device, both, or neither. Periprocedural adverse events included 2 episodes of cardiac tamponade and 1 episode of access site hematoma. These events point to the need for close postprocedure evaluation.
Indications. Our patients had AF and often contraindications to anticoagulation. Those that were able to take oral anticoagulants preferred the one‐time procedure instead of lifetime drug continuation. Systemic anticoagulation has the consequence of elevated bleeding risk leading to relative or absolute anticoagulation in up to 40% of patients with AF,12,13 frequent laboratory monitoring, and a significant amount of time outside therapeutic levels.14 New oral anticoagulants (e.g., dabigatran, rivaraoxaban, and apixaban) minimize food and drug interactions and do
Table 3. Procedural Data
LAA device Amplatzer PLAATO Watchman LAA device size (mm) Partial recaptures Full recaptures ASD/PFO Amplatzer Helex Starflex Figullaflex Premere PFO device size (mm) ASD device size (mm) ASD/PFO device size (mm) Fluoroscopy LAA proc (min) Fluoroscopy ASD proc (min) Fluoroscopy total (min)
LAA Then ASD/PFO (n ¼ 11)
Concurrent (n ¼ 3)
ASD/PFO Then LAA (n ¼ 3)
Total (n ¼ 17)
81.8% 18.2% 0% 22.1 4.7 27.3% 27.3%
0% 66.6% 33.3% 28.3 4.0 0% 0%
33.3% 66.6% 0% 30.0 3.5 33% 0%
58.8% 35.3% 5.9% 24.6 5.5 23.6% 11.8%
63.6% 9.1% 9.1% 9.1% 9.1% 21.6 3.2 17 3.5 19.3 4.0 12.6 8.3 10.8 11.9 23.4 8.3
100% 0% 0% 0% 0% 20.5 6.4 35 25.3 9.5
66.6% 33.3% 0% 0% 0% 15 27 15.6 26.4 16.3 16.1 7.1 7.4 3.1 23.5 6.8
70.6% 11.8% 5.9% 5.9% 5.9% 20.5 4.1 21.8 9.5 21.5 7.2
28.3 4.0
ASD, atrial septal defect; LAA, left atrial appendage; PFO, patent foramen ovale.
Vol. 27, No. 4, 2014
Journal of Interventional Cardiology
419
GAFOOR, ET AL. Table 4. Procedural Adverse Events LAA Then ASD/PFO (n ¼ 11) Tamponade Hematoma
Concurrent (n ¼ 3)
ASD then LAA (n ¼ 3)
Total (n ¼ 17)
0% 33%
0% 0%
11.7% 5.9%
18.2% 0%
During first procedure (LAA).
not require frequent monitoring, but there is a significant bleeding risk and drug intolerance, leading to discontinuance rates of 21% and 24% with dabigatran and rivaroxaban, respectively,15,16 and significantly higher costs compared to warfarin. LAA occlusion is effective at reducing the risk of stroke. The PLAATO multicenter registry of 111 patients with contraindication to anticoagulation reported a 10‐month stroke rate of 2.2% compared to CHADS2 predicted stroke rate of 6.3%.17 Similarly, the PLAATO North American Registry in 64 patients showed a 5‐year stroke rate of 3.8% compared to the expected 6.6%.18 The randomized, controlled PROTECT‐AF trial,19–21 showed noninferiority of LAA occlusion to warfarin for the primary efficacy end‐point (stroke, systemic embolism, or cardiovascular/unexplained death). Although the safety end‐point was worse in the WATCHMAN group (7.4% vs. 4.4% per 100 patients), there was a significant learning curve showing improvement in the latter half of patients.22 The PREVAIL study, designed to confirm the results of PROTECT‐AF, showed a safety rate better than the PROTECT AF trial (2.2% vs. 2.7%) and met the efficacy end‐point of late ischemic stroke and systemic embolism at 18 months (0.025 vs. 0.051 per 100 patient years). In addition, implant success was higher than PROTECT‐AF (95.1 vs. 90.9%).23 LAA closure by surgical techniques is also possible, but our goal was to attempt a percutaneous method.24,25 The mean age of our patients was 63 years. Percutaneous closure of an atrial septal defect has significant benefits to morbidity and mortality in different age groups including the elderly.26–28 After closure, patients show an improvement in symptoms, reduction in pulmonary artery pressure, and reverse remodeling of the right ventricle.26 Various international societies have come forward with specific guidelines regarding closure, as seen by the 2010 ESC27 and 2008 ACC/AHA28 Guidelines. Our patients fit within the European guidelines, where PA pressure two‐thirds systemic pressure and there was evidence of a significant L‐R
420
shunt. Of note, atrial tachyarrhythmias may occur early after intervention. However, our patients had longstanding AF and there is evidence to show that ASD closure after age of 40 years appears not to affect frequency of subsequent arrhythmia development.29,30 For patients with AF and PFO, the picture is the most controversial. Risk of developing stroke with a PFO is 1 in 1,000 per year.2,31 Recurrence rate of stroke in patients with prior cryptogenic stroke is approximately 2% annually (up to 15% if atrial septal aneurysm is present.32–34 Surgical closure of PFO in 193 patients showed a recurrence rate of stroke of 0.42% and TIA of 0.56%.35 Multiple methods of PFO closure exist, including device, radiofrequency, and stitch closure.36 A pooled analysis with 3,819 patients undergoing percutaneous PFO closure showed a recurrent annual stroke rate of 0.47% and recurrent TIA risk of 0.85%.35 Three multicenter randomized controls trials (CLOSURE,37 RESPECT,38 and PC39) have examined the use of PFO devices in patients with cryptogenic stroke and showed no significant reduction of stroke. However, multiple caveats exist, including high predilection for thrombus with the CLOSURE StarFlex PRO device (NMT Medical, Boston, MA, USA)40 and positive results for the RESPECT per‐protocol and as‐ treated groups. In addition, randomized trials of PFO closure face low event rates, require long follow‐up, and are confounded by patients who receive off‐label ASD devices to close PFOs.35 Patients with PFO closure had prior cryptogenic stroke, except for one patient. One patient in the “LAA occlusion followed by ASD/PFO closure,” or Group I, group had no prior stroke, but had PFO closure for migraines. Of note, a second patient in Group III had PFO closure for cryptogenic stroke and then developed atrial fibrillation. Evaluation of prior electrocardiograms showed no evidence of atrial fibrillation. After recognition of atrial fibrillation, he was placed on Vitamin K antagonists and had a subsequent stroke. The patient preferred to have left atrial appendage closure, which was performed 3 years later. For this
Journal of Interventional Cardiology
Vol. 27, No. 4, 2014
LEAVING NO HOLE UNCLOSED
patient, it is conceivable that he had occult atrial fibrillation prior to PFO closure, but this was only discovered after the fact. These patients illustrate the complex interactions between disease progression, medication effects, and patient preference that can influence prior decisions, which then, in turn, impact future treatment options. There are several limitations of this article. The sample size is not large enough to allow comparisons of the three techniques. All procedures were done based on patient preference and unwillingness/inability to take anticoagulation. The patients should be followed very closely after cessation of anticoagulation. In addition, this article does not go into efficacy of each approach, but only the procedural safety for the small numbers in each group. This article should therefore be used as a call for further research to determine the best technique, timing, and postprocedural anticoagulation strategy for patients with interatrial communication and AF. Therefore, it is not the purpose of the article to find a perfect strategy for all patients at all times, but rather to start a discussion of the strengths and drawbacks of all available strategies. Given the above results and limitations, we recommend a careful patient‐specific approach that utilizes the best of inconclusive data in this rarely studied but clinically significant population of patients with AF and interatrial communication. This requires a careful consideration of indication and technical issues. 1. Patients with secundum ASD and AF that are not candidates for anticoagulation: we recommend closure of the LAA and subsequent closure of the ASD as staged procedures or at the same sitting provided there is an interval of at least 10 minutes between the 2 procedures to allow for the small possibility of very early embolization of the LAA occluder device into the LA. It also is technically more feasible to close the LAA first through the ASD and then perform ASD closure. 2. Patients with AF (that are not candidates for anticoagulation) and PFO (without prior cryptogenic stroke): these patients should receive LAA occlusion alone. The risk of first stroke associated with the PFO is 40 years old. A randomized clinical trial. J Am Coll Cardiol 2001;38:2035–2042. Humenberger M, Rosenhek R, Gabriel H, et al. Benefit of atrial septal defect closure in adults: Impact of age. Eur Heart J 2011;32:553–560. Di Tullio M, Sacco RL, Gopal A, et al. Patent foramen ovale as a risk factor for cryptogenic stroke. Ann Intern Med 1992; 117:461–465. Bogousslavsky J, Garazi S, Jeanrenaud X, Aebischer N, Van Melle G. Stroke recurrence in patients with patent foramen ovale: The Lausanne Study. Lausanne Stroke with Paradoxal Embolism Study Group. Neurology 1996;46:1301–1305. Mas JL, Zuber M. Recurrent cerebrovascular events in patients with patent foramen ovale, atrial septal aneurysm, or both and cryptogenic stroke or transient ischemic attack. French Study Group on Patent Foramen Ovale and Atrial Septal Aneurysm. Am Heart J 1995;130:1083–1088. Mas JL, Arquizan C, Lamy C, et al. Recurrent cerebrovascular events associated with patent foramen ovale, atrial septal aneurysm, or both. N Engl J Med 2001;345:1740–1746. Tobis J, Shenoda M. Percutaneous treatment of patent foramen ovale and atrial septal defects. J Am Coll Cardiol 2012;60:1722– 1732. Kaplan AV. Burn it, stitch it, plug it … there are many ways to close a PFO: Insights into the challenges of medical device development. Catheter Cardiovasc Interv 2009;73:374–375. Furlan AJ, Reisman M, Massaro J, et al. Closure or medical therapy for cryptogenic stroke with patent foramen ovale. N Engl J Med 2012;366:991–999. Carroll JD, Saver JL, Thaler DE, et al. Closure of patent foramen ovale versus medical therapy after cryptogenic stroke. N Engl J Med 2013;368:1092–1100. Meier B, Kalesan B, Mattle HP, et al. Percutaneous closure of patent foramen ovale in cryptogenic embolism. N Engl J Med 2013;368:1083–1091. Taaffe M, Fischer E, Baranowski A, et al. Comparison of three patent foramen ovale closure devices in a randomized trial (Amplatzer versus CardioSEAL‐STARflex versus Helex occluder). Am J Cardiol 2008;101:1353–1358.
Journal of Interventional Cardiology
Vol. 27, No. 4, 2014
STRUCTURAL HEART DISEASE Leaving No Hole Unclosed: Left Atrial Appendage Occlusion in Patients Having Closure of Patent Foramen Ovale or Atrial Septal Defect SAMEER GAFOOR, M.D.,1 JENNIFER FRANKE, M.D.,2 PATRICK BOEHM,1 SIMON LAM, M.D.,1 STEFAN BERTOG, M.D.,1,3 LAURA VASKELYTE, M.D.,1 ILONA HOFMANN, M.D.,1 and HORST SIEVERT, M.D., PH.D.1 From the 1Cardiovascular Center Frankfurt, Frankfurt, Germany; 2University of Heidelberg, Heidelberg, Germany; and 3Minneapolis Veterans Affairs Hospital, Minneapolis, Minnesota
Objectives: To report procedural outcome of sequential occlusion of the left atrial appendage (LAA) and an interatrial septal communication and discuss possible indications. Background: There are some patients who may have indications for both closure of patent foramen ovale (PFO) or atrial septal defect (ASD), as well as closure of the LAA. The optimal procedural strategy is not known. Methods: A retrospective review of LAA and PFO/ASD cases at our center was performed. Demographic, echocardiographic, and procedural data were recorded. Results: Closure of LAA and then PFO/ASD (Group I: 11 patients), closure of both in the same setting (Group II: 3 patients), and closure of the PFO/ASD and then the LAA (Group III: 3 patients) was performed in a total of 17 patients. Average age was 63.5 9.8 years. Most patients were hypertensive with prior cerebrovascular event in 52.9% of patients. Procedural success was 100%. Procedural adverse events were 2 episodes of tamponade (in Group I after first LAA procedure) treated with pericardiocentesis and 1 access‐site hematoma (Group II) treated conservatively. Conclusions: Sequential (staged or during the same procedure) closure of the LAA and interatrial communications can be performed safely in a carefully selected patient population. This is also the first known report of LAA occlusion in patients with prior septal closure devices. (J Interven Cardiol 2014;27:414–422)
Introduction Via communication between the venous and systemic circulation, both patent foramen ovale and atrial septal defects may allow paradoxical embolism to occur. Hence both have been implicated as the cause of strokes.1–3 Moreover, atrial septal defects are associated with a higher incidence of atrial fibrillation (AF). Given the risk of paradoxical and AF‐related stroke with interatrial communication, anticoagulation is often recommended in patients with additional AF. However, anticoagulation with warfarin is associated with both extracranial and intracranial hemorrhage4
Address for reprints: Dr. Horst Sievert, Cardiovascular Center Frankfurt, Seckbacher Landstrasse 65, 60389 Frankfurt am Main, Germany. Fax: þ49‐69‐46031343; e‐mail: [email protected]
414
and up to 40% of patients with AF have contraindications to anticoagulation.5 Newer anticoagulants have similar bleeding rates compared to warfarin with high rates of discontinuation.6 For these patients, closure of the left atrial appendage may have merit. In patients with both interatrial communication and AF, closure of the interatrial communication and left atrial appendage may be warranted, especially in those with high bleeding risk or contraindications to anticoagulation. When a cerebrovascular event occurs in this setting, it is difficult to say whether the cause is the interatrial communication or thrombus from AF. Little data exist evaluating the safety of closing both the LAA and an interatrial communication, with only 2 reports in the literature.7,8 The purpose of this study was to evaluate feasibility and safety of percutaneous closure of both the LAA and interatrial septal communication.
Journal of Interventional Cardiology
Vol. 27, No. 4, 2014
LEAVING NO HOLE UNCLOSED
Methods This was a retrospective review analysis of consecutive cases of both percutaneous interatrial septal communication repair (ASD/PFO) and left atrial appendage occlusion from 2002 to 2013. The patients were divided into 3 groups based on whether the patients had LAA occlusion prior, in the same setting, or after ASD/PFO closure. Key demographic, echocardiographic, and procedural data were collected for each patient. Follow‐up was performed using a combination of follow‐up visits as well as phone and mail survey assessment. Left Atrial Appendage Occlusion. Patients were selected for left atrial appendage closure based on a history of AF and inability to take anticoagulation or patient desire to be off of anticoagulation. Percutaneous left atrial appendage closure was performed with femoral venous access and transseptal puncture. Size of device was determined based on measurements of LAA from angiography and echocardiogram. Device was placed and sheath was removed. Patients had follow‐up echocardiography the next day and were discharged on a regimen of aspirin indefinitely and clopidogrel for 3– 6 months. Atrial Septal Defect/Patent Foramen Ovale Closure. Atrial septal defect was closed according to guidelines for congenital heart disease. Patent foramen ovale was closed for cryptogenic stroke. Via standard femoral venous access, balloon sizing was performed to evaluate the width of the atrial septal defect or the length of the patent foramen ovale tunnel. Concurrent transesophageal echocardiogram was used to evaluate the interatrial septum for other defects. The defect was percutaneously closed using an appropriate PFO or ASD occluder and then evaluated with transesophageal echocardiogram and angiography for residual shunt. Group 1: LAA Occlusion Followed by Interatrial Communication Closure in Separate Settings. For this group of patients, the procedure was performed similar to the techniques mentioned above. The puncture site for LAA occlusion may be through the existing interatrial communication (although some operators call for a new puncture site). Clinical setting dictated the follow‐up time period between the first intervention (LAA occlusion) and the second intervention (interatrial communication closure). During the interatrial communication closure, care was taken to
Vol. 27, No. 4, 2014
place the wire in the left upper pulmonary vein without disturbing the LAA occlusion device. This requires a good understanding of the left upper pulmonary vein and LAA anatomical relationship. Group II: LAA Occlusion and Interatrial Communication Closure in the Same Setting. The LAA occlusion was performed first, followed by interatrial communication closure. In this setting, the LAA occlusion was performed through the interatrial communication (Fig. 3). It is our protocol to wait 5– 10 minutes after LAA occlude placement before closing the PFO/ASD. This maintains access to the left atrium in case of early LAA occluder embolization. From our experience, this may occur even with good placement, adequate compression, and follow‐up angiography showing trace or minimal leak. One limitation of this method is that in the rare case of infection after the procedure, it is difficult to know whether the infected device is the LAA occlusion device, the interatrial septum closure device, both, or neither. Group III: PFO Closure Followed by LAA Occlusion in Separate Settings. After interatrial communication closure, the patient may develop atrial fibrillation and be a candidate for LAA closure. In this case, the procedure is more challenging as there is a device at the septum. It is our practice to attempt a puncture immediately inferior to the device. This is accomplished by observing the septum in the 30° and 90° view and puncturing inferior to the device and posterior to the aorta (Fig. 1). However, anterior puncture may also be used (Fig. 2), as long as the aorta is far away from the puncture site. Another option is to cross through the device, which is possible with some devices.9,10
Results Baseline demographic data are shown in Table 1. A total of 17 patients (mean age 63.5 years) were identified. Eleven patients first underwent closure of the LAA and then ASD/PFO (Group I), in 3 both procedures performed in the same setting (Group II), and in 3 patients ASD/PFO closure was performed first followed by LAA occlusion (Group III). In Group I, the time between LAA occlusion and ASD/PFO closure was 8.4 6.8 months, while for group III the time between ASD/PFO closure and LAA occlusion was 22.8 20.3 months. The majority of patients had paroxysmal AF, with a trend toward more permanent
Journal of Interventional Cardiology
415
GAFOOR, ET AL.
Figure 1. Placement of LAA occluder in a patient with prior ASD closure. The procedure began with a transseptal puncture inferior to the ASD occluder in the LAO view (A), followed by advancement of a pigtail catheter to take measurements of the LAA (B). This was then exchanged (C, D) for a large sheath and an LAA occluder was advanced (E) and deployed (F). At deployment one can see the relationship between the LAA occluder and ASD.
AF in Group III. One patient in Group I had cardioversion prior to the procedure. Pulmonary vein isolation or MAZE procedures were not performed in any of the patients. More than half of the patients had a history of thromboembolism, except for Group III in which only 1 patient had a history of stroke. CHADS2‐ Score was 2 across all Groups, and the mean CHA2DS2‐VASC‐Score was at least 3.9. Four of the 17 patients had previous bleeding complications on anticoagulation, and 1 had an allergic reaction. Therefore, just under half of patients were on anticoagulation, with the exception of Group III where 2 of 3 patients were on anticoagulation. Detailed echocardiographic data are shown in Table 2. The patients had normal left ventricular ejection fraction except for 1 patient. Approximately half of the patients had an atrial septal defect and the other half a patent foramen ovale. One patient in Group I had both. All patients with PFO had cryptogenic stroke except for 1 patient with PFO in Group I who had closure secondary to migraines. Another patient with PFO and cryptogenic
416
stroke in Group III was found to have subsequent AF that was not present at the time of first procedure. Table 3 shows procedural data for both procedures. Procedural success was 100%. For LAA occlusion, the Amplatzer device (ACP) was used in the vast majority of cases, followed by the PLAATO occluder. For patients with LAA occlusion followed by ASD/PFO closure, the ASD/PFO was used for access for the LAA occlusion. Group I had the most partial (3) and full (3) recaptures, of which 2 occurred in the same patient. For interatrial communication closure, the Amplatzer ASD/ PFO device was used in the vast majority (70.6%) of cases, followed by the Helex closure device (11.8%). Mean total fluoroscopy time was longer for the combined procedure (Group II) than individual procedures (Groups I and III). Procedural adverse events are shown in Table 4. These include pericardial effusion in 2 patients in Group I (treated successfully with pericardiocentesis) and a hematoma in 1 patient in Group II. Both adverse events in Group I happened in relation to the first procedure (LAA occlusion).
Journal of Interventional Cardiology
Vol. 27, No. 4, 2014
LEAVING NO HOLE UNCLOSED
Discussion
Figure 2. Transseptal procedure in patient with prior PFO device. Transesophageal echocardiogram at mid‐esophageal level showing transseptal procedure in patient with prior PFO device (left arrow). In this patient, the crossing point for the transseptal needle was in the anterior part of the septum (right arrow).
PFO (with cryptogenic stroke) and ASD are both risk factors for stroke,1,3 which is only increased by the presence of AF. One strategy in selected patients is closure of both the interatrial communication and left atrial appendage. To our knowledge, this is the first case series of such patients with both interatrial septal closure and left atrial appendage occlusion. There are both technical and indication‐based issues that would drive the order and necessity of both these procedures in the same patient. Technical Issues. The technique behind interatrial septal defect closure and LAA occlusion has been previously defined. This article highlights the ability to have a strategy where both procedures can be done in the same setting or in subsequent procedures. When closing the LAA first, the procedure is straight forward. When present, the PFO or ASD can be used for transseptal access. Some operators have called for a different and new transseptal puncture site for LAA
Figure 3. Placement of the LAA occluder and PFO closure device in the same setting. After crossing the septum through the PFO, balloon sizing was used to measure the orifice diameter (A). Next, a pigtail was placed in the LAA (B) and used to exchange for a delivery sheath to allow LAA occluder placement (C). Next, the same device sheath was used for PFO occluder delivery and deployment (D, E). The final position of the LAA occluder and PFO closure device is seen in the AP view (F).
Vol. 27, No. 4, 2014
Journal of Interventional Cardiology
417
GAFOOR, ET AL. Table 1. Baseline Characteristics
Age at first intervention (years) Time between interventions (months) Height (cm) Weight (kg) Hypertension Hyperlipidemia Diabetes Migraine Coronary artery disease Myocardial infarction Congestive heart failure NYHA class Peripheral vascular disease Deep vein thrombosis Atrial fibrillation Paroxysmal Persistent Permanent Cardioversion EP ablation MAZE procedure Previous TIA Previous CVA History of thromboembolism (TIA/CVA) CHADS2 CHADS2‐VASc Previous bleeding complication on anticoagulation (number of patients) GI bleeding Liver disease Other contraindication for vitamin K antagonist (VKA) Medications at time of first intervention Aspirin Plavix VKA
LAA Then ASD/PFO (11 Patients)
Concurrent (3 Patients)
ASD/PFO Then LAA (3 Patients)
Total (17 Patients)
63.4 8.8 8.44 6.8 169.1 10.4 78.8 17.7 90.9% 25% 9.1% 9.1% 27.3% 0% 27.3% 1.2 0.8 0% 0%
59 5.7 0 172 5.7 82.3 17.3 100% 0% 33% 0% 33% 0% 0% 0 0% 33.3%
63.3 16.7 22.8 20.3 181 8.5 93 22.6 100% 66.7% 66.7% 33.3% 66.7% 66.6% 66.7% 1.7 0.6 33.3% 33.3%
63.5 9.8 9.5 11.4 171 9.8 80.5 17.2 94.1% 35.3% 23.5% 11.7% 35.3% 11.7% 29.4% 1.1 0.8 6% 11.8%
63.6% 18.2% 18.2% 8.7% 0% 0% 36% 45% 55% 2.4 1.0 3.7 1.2 1 (hematoma, epistaxis) 0% 0% 0
66.6% 33.3% 0% 0% 0% 0% 50% 66.7% 66.7% 31 41 1 (retinal)
52.9% 23.5% 23.5% 0% 0% 0% 35.3% 47.1% 52.9% 2.6 0.9 3.9 1.1 4
0% 0% 0
0% 33.3% 66.6% 0% 0% 0% 0% 33.3% 33.3% 30 4.7 0.6 2 (gastrointestinal, epistaxis) 0% 0% 1 (allergic reaction)
18.2% 9.1% 36.4%
33.3% 0% 33.3%
0% 0% 66.6%
17.6% 5.9% 41.2%
0% 0%
CVA, cardiovascular accident; EP, electrophysiology; GI, gastrointestinal; NYHA, New York Heart Association; TIA, transient ischemic attack.
occlusion.11 In our experience, a new septal puncture is usually not necessary and the septum can be safely crossed through the PFO or ASD. Second, care must be taken to not dislodge the prior placed LAA device when inserting the PFO/ASD sheath. This requires a good understanding of the left upper pulmonary vein and left atrial appendage anatomy. When interatrial communication closure is followed by LAA occlusion, the procedure is more challenging. One may cross the septum around the device. This can be done in various areas; however, it is our practice to attempt a puncture immediately inferior to the device when possible
418
(Fig. 1). However, anterior puncture may also be used (Fig. 2). The other option includes crossing through the device, which from our experience is possible with some devices.9,10 To our knowledge, this is the first report of LAA occlusion in patients with prior septal closure devices. When both procedures are done in the same setting, it is easier to use the interatrial communication for advancement of the LAA sheath and device to be followed by ASD or PFO closure (Fig. 3). However, significant issues with this approach are also present. First, closure of the interatrial septum shortly after LAA occlusion would preclude quick access to the
Journal of Interventional Cardiology
Vol. 27, No. 4, 2014
LEAVING NO HOLE UNCLOSED Table 2. Echocardiographic Data
Ejection fraction (%) Valvular disease Mitral regurgitation (presence of) Interatrial communication ASD PFO Left atrial appendage multiple lobes (%) Maximal measurement at ostium (mm)
LAA Then ASD/PFO (n ¼ 11)
Concurrent (n ¼ 3)
ASD/PFO Then LAA (n ¼ 3)
Total (n ¼ 17)
67.9 10.7
64 3.6
49.3 9.8
63.1 11.7
63.6%
100%
66.6%
70.6%
54.4% 54.4%
33.3% 66.6%
66.6% 33.3%
52.9% 52.9%
45.5% 17.8 5.5
0% 22.1 1.4
33.3% 23.4 1.5
35.3% 19.8 4.9
One patient in this group had a hybrid defect of both ASD and PFO.
left atrium in case of device embolization. This could be avoided by placing the LAA occluder and then postponing PFO/ASD closure for 10–15 minutes to see if the device is prone to early embolization. Second, in the rare case of infection after the procedure, it is difficult to know whether the infected device is the LAA occlusion device, the interatrial septum closure device, both, or neither. Periprocedural adverse events included 2 episodes of cardiac tamponade and 1 episode of access site hematoma. These events point to the need for close postprocedure evaluation.
Indications. Our patients had AF and often contraindications to anticoagulation. Those that were able to take oral anticoagulants preferred the one‐time procedure instead of lifetime drug continuation. Systemic anticoagulation has the consequence of elevated bleeding risk leading to relative or absolute anticoagulation in up to 40% of patients with AF,12,13 frequent laboratory monitoring, and a significant amount of time outside therapeutic levels.14 New oral anticoagulants (e.g., dabigatran, rivaraoxaban, and apixaban) minimize food and drug interactions and do
Table 3. Procedural Data
LAA device Amplatzer PLAATO Watchman LAA device size (mm) Partial recaptures Full recaptures ASD/PFO Amplatzer Helex Starflex Figullaflex Premere PFO device size (mm) ASD device size (mm) ASD/PFO device size (mm) Fluoroscopy LAA proc (min) Fluoroscopy ASD proc (min) Fluoroscopy total (min)
LAA Then ASD/PFO (n ¼ 11)
Concurrent (n ¼ 3)
ASD/PFO Then LAA (n ¼ 3)
Total (n ¼ 17)
81.8% 18.2% 0% 22.1 4.7 27.3% 27.3%
0% 66.6% 33.3% 28.3 4.0 0% 0%
33.3% 66.6% 0% 30.0 3.5 33% 0%
58.8% 35.3% 5.9% 24.6 5.5 23.6% 11.8%
63.6% 9.1% 9.1% 9.1% 9.1% 21.6 3.2 17 3.5 19.3 4.0 12.6 8.3 10.8 11.9 23.4 8.3
100% 0% 0% 0% 0% 20.5 6.4 35 25.3 9.5
66.6% 33.3% 0% 0% 0% 15 27 15.6 26.4 16.3 16.1 7.1 7.4 3.1 23.5 6.8
70.6% 11.8% 5.9% 5.9% 5.9% 20.5 4.1 21.8 9.5 21.5 7.2
28.3 4.0
ASD, atrial septal defect; LAA, left atrial appendage; PFO, patent foramen ovale.
Vol. 27, No. 4, 2014
Journal of Interventional Cardiology
419
GAFOOR, ET AL. Table 4. Procedural Adverse Events LAA Then ASD/PFO (n ¼ 11) Tamponade Hematoma
Concurrent (n ¼ 3)
ASD then LAA (n ¼ 3)
Total (n ¼ 17)
0% 33%
0% 0%
11.7% 5.9%
18.2% 0%
During first procedure (LAA).
not require frequent monitoring, but there is a significant bleeding risk and drug intolerance, leading to discontinuance rates of 21% and 24% with dabigatran and rivaroxaban, respectively,15,16 and significantly higher costs compared to warfarin. LAA occlusion is effective at reducing the risk of stroke. The PLAATO multicenter registry of 111 patients with contraindication to anticoagulation reported a 10‐month stroke rate of 2.2% compared to CHADS2 predicted stroke rate of 6.3%.17 Similarly, the PLAATO North American Registry in 64 patients showed a 5‐year stroke rate of 3.8% compared to the expected 6.6%.18 The randomized, controlled PROTECT‐AF trial,19–21 showed noninferiority of LAA occlusion to warfarin for the primary efficacy end‐point (stroke, systemic embolism, or cardiovascular/unexplained death). Although the safety end‐point was worse in the WATCHMAN group (7.4% vs. 4.4% per 100 patients), there was a significant learning curve showing improvement in the latter half of patients.22 The PREVAIL study, designed to confirm the results of PROTECT‐AF, showed a safety rate better than the PROTECT AF trial (2.2% vs. 2.7%) and met the efficacy end‐point of late ischemic stroke and systemic embolism at 18 months (0.025 vs. 0.051 per 100 patient years). In addition, implant success was higher than PROTECT‐AF (95.1 vs. 90.9%).23 LAA closure by surgical techniques is also possible, but our goal was to attempt a percutaneous method.24,25 The mean age of our patients was 63 years. Percutaneous closure of an atrial septal defect has significant benefits to morbidity and mortality in different age groups including the elderly.26–28 After closure, patients show an improvement in symptoms, reduction in pulmonary artery pressure, and reverse remodeling of the right ventricle.26 Various international societies have come forward with specific guidelines regarding closure, as seen by the 2010 ESC27 and 2008 ACC/AHA28 Guidelines. Our patients fit within the European guidelines, where PA pressure two‐thirds systemic pressure and there was evidence of a significant L‐R
420
shunt. Of note, atrial tachyarrhythmias may occur early after intervention. However, our patients had longstanding AF and there is evidence to show that ASD closure after age of 40 years appears not to affect frequency of subsequent arrhythmia development.29,30 For patients with AF and PFO, the picture is the most controversial. Risk of developing stroke with a PFO is 1 in 1,000 per year.2,31 Recurrence rate of stroke in patients with prior cryptogenic stroke is approximately 2% annually (up to 15% if atrial septal aneurysm is present.32–34 Surgical closure of PFO in 193 patients showed a recurrence rate of stroke of 0.42% and TIA of 0.56%.35 Multiple methods of PFO closure exist, including device, radiofrequency, and stitch closure.36 A pooled analysis with 3,819 patients undergoing percutaneous PFO closure showed a recurrent annual stroke rate of 0.47% and recurrent TIA risk of 0.85%.35 Three multicenter randomized controls trials (CLOSURE,37 RESPECT,38 and PC39) have examined the use of PFO devices in patients with cryptogenic stroke and showed no significant reduction of stroke. However, multiple caveats exist, including high predilection for thrombus with the CLOSURE StarFlex PRO device (NMT Medical, Boston, MA, USA)40 and positive results for the RESPECT per‐protocol and as‐ treated groups. In addition, randomized trials of PFO closure face low event rates, require long follow‐up, and are confounded by patients who receive off‐label ASD devices to close PFOs.35 Patients with PFO closure had prior cryptogenic stroke, except for one patient. One patient in the “LAA occlusion followed by ASD/PFO closure,” or Group I, group had no prior stroke, but had PFO closure for migraines. Of note, a second patient in Group III had PFO closure for cryptogenic stroke and then developed atrial fibrillation. Evaluation of prior electrocardiograms showed no evidence of atrial fibrillation. After recognition of atrial fibrillation, he was placed on Vitamin K antagonists and had a subsequent stroke. The patient preferred to have left atrial appendage closure, which was performed 3 years later. For this
Journal of Interventional Cardiology
Vol. 27, No. 4, 2014
LEAVING NO HOLE UNCLOSED
patient, it is conceivable that he had occult atrial fibrillation prior to PFO closure, but this was only discovered after the fact. These patients illustrate the complex interactions between disease progression, medication effects, and patient preference that can influence prior decisions, which then, in turn, impact future treatment options. There are several limitations of this article. The sample size is not large enough to allow comparisons of the three techniques. All procedures were done based on patient preference and unwillingness/inability to take anticoagulation. The patients should be followed very closely after cessation of anticoagulation. In addition, this article does not go into efficacy of each approach, but only the procedural safety for the small numbers in each group. This article should therefore be used as a call for further research to determine the best technique, timing, and postprocedural anticoagulation strategy for patients with interatrial communication and AF. Therefore, it is not the purpose of the article to find a perfect strategy for all patients at all times, but rather to start a discussion of the strengths and drawbacks of all available strategies. Given the above results and limitations, we recommend a careful patient‐specific approach that utilizes the best of inconclusive data in this rarely studied but clinically significant population of patients with AF and interatrial communication. This requires a careful consideration of indication and technical issues. 1. Patients with secundum ASD and AF that are not candidates for anticoagulation: we recommend closure of the LAA and subsequent closure of the ASD as staged procedures or at the same sitting provided there is an interval of at least 10 minutes between the 2 procedures to allow for the small possibility of very early embolization of the LAA occluder device into the LA. It also is technically more feasible to close the LAA first through the ASD and then perform ASD closure. 2. Patients with AF (that are not candidates for anticoagulation) and PFO (without prior cryptogenic stroke): these patients should receive LAA occlusion alone. The risk of first stroke associated with the PFO is 40 years old. A randomized clinical trial. J Am Coll Cardiol 2001;38:2035–2042. Humenberger M, Rosenhek R, Gabriel H, et al. Benefit of atrial septal defect closure in adults: Impact of age. Eur Heart J 2011;32:553–560. Di Tullio M, Sacco RL, Gopal A, et al. Patent foramen ovale as a risk factor for cryptogenic stroke. Ann Intern Med 1992; 117:461–465. Bogousslavsky J, Garazi S, Jeanrenaud X, Aebischer N, Van Melle G. Stroke recurrence in patients with patent foramen ovale: The Lausanne Study. Lausanne Stroke with Paradoxal Embolism Study Group. Neurology 1996;46:1301–1305. Mas JL, Zuber M. Recurrent cerebrovascular events in patients with patent foramen ovale, atrial septal aneurysm, or both and cryptogenic stroke or transient ischemic attack. French Study Group on Patent Foramen Ovale and Atrial Septal Aneurysm. Am Heart J 1995;130:1083–1088. Mas JL, Arquizan C, Lamy C, et al. Recurrent cerebrovascular events associated with patent foramen ovale, atrial septal aneurysm, or both. N Engl J Med 2001;345:1740–1746. Tobis J, Shenoda M. Percutaneous treatment of patent foramen ovale and atrial septal defects. J Am Coll Cardiol 2012;60:1722– 1732. Kaplan AV. Burn it, stitch it, plug it … there are many ways to close a PFO: Insights into the challenges of medical device development. Catheter Cardiovasc Interv 2009;73:374–375. Furlan AJ, Reisman M, Massaro J, et al. Closure or medical therapy for cryptogenic stroke with patent foramen ovale. N Engl J Med 2012;366:991–999. Carroll JD, Saver JL, Thaler DE, et al. Closure of patent foramen ovale versus medical therapy after cryptogenic stroke. N Engl J Med 2013;368:1092–1100. Meier B, Kalesan B, Mattle HP, et al. Percutaneous closure of patent foramen ovale in cryptogenic embolism. N Engl J Med 2013;368:1083–1091. Taaffe M, Fischer E, Baranowski A, et al. Comparison of three patent foramen ovale closure devices in a randomized trial (Amplatzer versus CardioSEAL‐STARflex versus Helex occluder). Am J Cardiol 2008;101:1353–1358.
Journal of Interventional Cardiology
Vol. 27, No. 4, 2014
