Transseptal catheterization is frequently performed to

Transcription

1 The Incidence and Long-Term Clinical Outcome of Iatrogenic Atrial Septal Defects Secondary to Transseptal Catheterization With a 12F Transseptal Sheath Sheldon M. Singh, MD; Pamela S. Douglas, MD, MACC, FAHA; Vivek Y. Reddy, MD, FACC Background Studies assessing the presence of a residual iatrogenic atrial septal defect (iasd) after transseptal catheterization with 8F transseptal sheaths have suggested that the majority of these iasds close within 6 months. However, these studies have been limited by small patient numbers and short follow-up. Additionally, there are a number of novel catheter procedures in interventional cardiology and electrophysiology that use larger transseptal sheaths. The objective of this study was to assess the incidence of and complications associated with iasds in a large cohort of patients undergoing transseptal catheterization with a 12F transseptal sheath. Methods and Results Patients (n 253) without a preexisting interatrial shunt undergoing WATCHMAN implantation as part of the PROTECT AF study were included in this current study. Patients underwent transesophageal echocardiography with echo-contrast immediately after the procedure and at 45 days and 6 months and 12 months. Eighty-seven percent of patients had an iasd immediately after the procedure, the majority of which sealed by 6 months (incidence of iasd, 34% at 45 days, 11% at 6 months, 7% at 12 months). Whereas the majority of iasds were 3 mm in diameter immediately after the procedure, the minority of iasd were 3 mm during the follow-up period. Additionally, interatrial shunting was predominantly left-to-right when an iasd was present. There was no significant difference in the rate of stroke and/or systemic embolism during the follow-up period in patients with or without iasd. Conclusions Transseptal catheterization procedures with a large-diameter transseptal sheath have a high spontaneous closure rate of iasds that is not associated with an increased rate of stroke/systemic embolization during long-term follow-up. (Circ Arrhythm Electrophysiol. 2011;4: ) Key Words: atrial septal defect transseptal catheterization atrial fibrillation Transseptal catheterization is frequently performed to access the left heart during percutaneous cardiac procedures. The increasing complexity of these procedures has resulted in the use of larger-diameter transseptal sheaths. For example, a 12F transseptal sheath (with a 14F outer diameter) is used during implantation of the WATCHMAN left atrial appendage (LAA) closure device, 1 pulmonary vein isolation procedures with the Endoscopic Laser balloon ablation catheter system, 2 and use of the Hansen robotic navigation system. 3 The use of such large-diameter sheaths has raised concerns of persistent iatrogenic atrial septal defects (iasds), which may be associated with right-to-left cardiac shunting and paradoxical embolic events. Current studies assessing the incidence of iasds after transseptal catheterization suggest that the majority of iasds close by 6 months. 4 6 However, these studies are limited by small sample sizes and short follow-up periods, which may prevent an accurate estimation of the incidence and complications associated with iasd. In addition, these data were largely obtained after transseptal puncture using 8F sheaths. Clinical Perspective on p 171 The PROTECT AF study was a prospective, multicenter, randomized, controlled trial that evaluated the safety and efficacy of percutaneous LAA closure versus warfarin therapy in patients with nonvalvular atrial fibrillation. 1 Patients randomly assigned to the WATCHMAN device implantation subsequently underwent serial transesophageal echocardiography (TEE) imaging, thereby permitting an assessment of the natural history of iasds secondary to the use of a large-diameter transseptal sheath. We report on the incidence Received July 29, 2010; accepted December 20, From the Cardiac Arrhythmia Service (S.M.S., V.Y.R.), Mount Sinai School of Medicine, New York, NY; and Duke Clinical Research Institute (P.S.D.), Durham, NC. Correspondence to Vivek Y. Reddy, MD, Mount Sinai School of Medicine, Zena and Michael A. Weiner Cardiovascular Institute, One Gustave L. Levy Place, Box 1030, New York, NY vivek.reddy@mountsinai.org 2011 American Heart Association, Inc. Circ Arrhythm Electrophysiol is available at DOI: /CIRCEP

2 Singh et al iasd After Transseptal Puncture With a 12F Sheath 167 and clinical outcomes associated with iasds in a large subgroup of patients without a preexisting interatrial shunt who underwent transseptal catheterization with a single 12F transseptal sheath. Methods The PROTECT AF trial was a prospective, randomized, controlled trial evaluating the efficacy of LAA closure with the WATCHMAN device compared with warfarin therapy in patients age 18 years and older with a history of paroxysmal, persistent, or permanent nonvalvular atrial fibrillation and CHADS 2 score of 1 or more. The details and results of the study have been previously published. 1 Four hundred sixty-three patients were assigned to receive LAA closure during the randomization process, 14 of whom device implantation was not attempted and 41 of whom device implantation was attempted but not possible. Additionally, the device was successfully implanted in 77 non randomly assigned roll-in patients, for a total of 485 patients who underwent device implantation in this trial. Exclusion Criteria Patients with a patent foramen ovale, atrial septal aneurysm (excursion and/or length 15 mm), and/or large shunt (within 3 beats and accompanied by a substantial passage of bubbles) were excluded from the PROTECT AF study. Additionally, patients without a baseline echo result, those with any interatrial shunt before WATCHMAN device implantation, or those with missing echo results available at a follow-up visits were excluded from the current subgroup of patients reported in the present study. TEE Imaging All patients underwent TEE imaging at baseline, intraoperatively, and at 45 days and 6 and 12 months, using commercially available TEE imaging equipment with at least a 5.0-MHz multiplane transducer. Multiple images of the interatrial septum were obtained both with and without color Doppler. In all patients, agitated saline was injected to assess for the presence of an interatrial shunt. The size and direction of the detected interatrial shunt was documented. If present, the length and excursion of an interatrial septal aneurysm were noted. Shunt fractions (Qp/Qs) were not routinely determined. All images were obtained for at least 5 cardiac cycles in each 2D view, or 10 cardiac cycles if the patient was not in sinus rhythm. Images and measurements were reviewed by an independent Core Echocardiography Laboratory (COResearch, Durham, NC). Transseptal Puncture Transseptal puncture was frequently performed with a long 8.5F SL-1 type vascular sheath that was subsequently exchanged over a guide wire for the 12F transseptal sheath (with a 14F outer diameter) used for delivering the WATCHMAN device (Figure 1). Transseptal punctures were generally performed with TEE guidance. Typically, transseptal punctures were placed slightly anterior to the typical puncture site used during atrial fibrillation ablation procedures. Follow-Up After successful device implantation, all patients remained on warfarin for a minimum of 45 days to permit device endothelialization. Warfarin therapy was discontinued at 45 days if TEE imaging demonstrated complete exclusion of the LAA or residual peridevice flow with jet width 5 mm, at which point daily clopidogrel (75 mg) and aspirin (81 to 325 mg) was administered an additional 4.5 months, followed by daily aspirin indefinitely. Figure 1. The 14F outer diameter WATCHMAN transseptal sheath. All adverse events (including neurological and other embolic events) were reviewed and adjudicated by an independent clinical events committee. Statistical Analysis Continuous variables are reported as mean SD. Categorical variables are reported in frequencies (percentages). Significant differences between patients with and without an iasd at 12 months were determined with univariate analysis of the 2 groups using the Student t test for continuous and the 2 test for categorical variables. The relationship between the occurrence of ischemic strokes/systemic embolism and the presence of an iasd was assessed using a Cox proportional hazards model with adjustment made for CHADS 2 score. All statistical analysis was performed with SAS software version 9.2 (SAS Institute Inc, Cary, NC). Probability values 0.05 were considered statistically significant. Ethics All patients provided written informed consent before participating in the study. The study was reviewed by the US Food and Drug Administration and was approved by each participating center s institutional review board. The study sponsor had no role in the interpretation of these data or writing of the final report. Results Four hundred eighty-five patients underwent implantation of the WATCHMAN device in the PROTECT AF study. Of these, 449 patients had baseline atrial shunt status available. We report on the cohort of 253 patients without a prior interatrial shunt who subsequently underwent implantation of the WATCHMAN device using the single 12F transseptal access sheath and who had complete follow-up for atrial shunt status. Excluded from this report are those without evidence of an absence of interatrial shunt before WATCHMAN device implantation (n 76 or those missing echo results at one or more follow-up visits (n 120). The average follow-up per patient in this cohort was 29 months (range, 11 to 55 months). Frequency of iasds During the Follow-Up Period TEE imaging was performed and reviewed by the core laboratory for intervals after the procedure and at 45 days, 6 months, and 12 months in 211 (83%), 234 (92%), 218 (86%) and 208 (82%) patients, respectively. Table 1

3 168 Circ Arrhythm Electrophysiol April 2011 Table 1. Proportion of Patients, Size, and Direction of Interatrial Shunt Immediately, 45 days, and 6 and 12 Months After WATCHMAN Implantation Patients With Shunt Size, % Direction of Shunt Shunt, n (%) 3 mm 3 mm Unknown Left-to-Right Bidirectional Right-to-Left Unknown Postprocedure 183 (87%) 69 (38%) 111 (61%) 3 (2%) 164 (90%) 9 (5%) 7 (4%) 3 (2%) 45 Days 80 (34%) 56 (70%) 22 (28%) 2 (3%) 65 (81%) 8 (10%) 5 (6%) 2 (3%) 6 Months 25 (11%) 17 (68%) 6 (24%) 2 (8%) 22 (88%) 1 (4%) 2 (8%) 0 (0%) 12 Months 14 (7%) 7 (50%) 4 (29%) 3 (21%) 7 (50%) 3 (21%) 4 (29%) 0 (0%) summarizes the proportion of patients with iasds and the size and direction of iasd shunting documented on TEE imaging immediately after the procedure and at 45 days and 6 and 12 months. An iasd was detected in 87% of patients immediately after the procedure, with this value decreasing to 34% at 45 days, 11% at 6 months, and 7% at 1 year. Although the majority of patients with iasds immediately after the procedure had an iasd 3 mm (Figure 2), only a minority of patients had an iasd 3 mm from 45 days onward. Left-to-right shunting was predominant when an iasd shunt was noted (Figure 3). Factors Associated With Postprocedural iasd Table 2 summarizes demographic, anatomic, and procedural characteristics for patients with and without an iasd noted immediately after the procedure. There was no association between factors such as age, CHADS 2 score, LA size, procedure time, and the presence of a postprocedural iasd. Of note, the LA pressure measured on the day of WATCHMAN device implantation was elevated and significantly higher in patients with an iasd compared with patients without an iasd ( mm Hg versus mm Hg, P 0.011). Although there was no significant difference in the proportion of patients with mitral regurgitation, the average degree of mitral regurgitation was not available for each group. Embolic Events During the Follow-Up Period Nine ischemic strokes/systemic embolism occurred during the follow-up period in this cohort: 5 occurred in patients with an immediate postprocedure iasd and 4 in patients without a postprocedure iasd. Six of the ischemic strokes/ systemic embolism occurred after 1 year of follow-up; 1 occurred in a patient with an iasd present on the 12-month TEE and 5 in patients without an iasd present on the 12-month TEE. Table 3 summarizes the event rate per patient-years of follow-up and hazard ratios with confidence intervals (unadjusted and adjusted for CHADS 2 score) for patients with and without an iasd detected immediately after the procedure, at 12 months, and the event rate per patient-years of follow-up for events occurring after 12 months. There was no significant increase in the risk of stroke/systemic embolism in patients with an iasd after the procedure or at 12 months. Discussion In experienced hands, transseptal catheterization can be performed with minimal complications 7 ; however, acute procedure-related iasd are inevitable. This study assessed the long-term incidence of iasds in a large cohort of patients undergoing transseptal catheterization with a 12F inner/14f outer diameter transseptal sheath. The main findings of this study are (1) the incidence of iasd is high (87%) immediately after transseptal catheterization, (2) the majority of iasds resolve within 45 days of the procedure, with less than 10% of the patients having a residual iasd at 12 months, (3) right-to-left shunting is rare when an iasd is present, and (4) patients with an iasd did not have a significant increase in the risk of stroke/systemic embolism compared with patients without an iasd. Figure 2. Atrial septal defect 3 mm in size noted immediately after WATCHMAN implantation. Color Doppler flow across the atrial septum immediately after WATCHMAN implantation demonstrating a 3.7-mm iasd with a left-to-right shunt. LA indicates left atrium; RA, right atrium. Figure 3. Left-to-right interatrial shunting in a patient 6 months after WATCHMAN implantation. Color Doppler flow across the atrial septum 6 months after WATCHMAN implantation demonstrating a small ( 3 mm) iasd with a left-to-right shunt. LA indicates left atrium; RA, right atrium.

4 Singh et al iasd After Transseptal Puncture With a 12F Sheath 169 Table 2. Demographic, Clinical, and Procedural Characteristics for Patients With and Without a Postprocedural Interatrial Shunt Postprocedure Shunt Present Postprocedure Shunt Absent P Value Age, y, mean SD Female sex, n (%) 53/183 (29.0%) 21/70 (30.0%) Hypertension, n (%) 170/183 (92.9%) 60/70 (85.7%) Diabetes, n (%) 36/183 (19.7%) 21/70 (30.0%) Prior TIA/stroke, n (%) 33/183 (18.0%) 12/70 (17.1%) CHADS 2 score, mean SD Paroxysmal atrial fibrillation, 79/183 (43.2%) 26/70 (37.1%) n (%) On warfarin at 12 mo, n (%) 13/183 (7.1%) 6/70 (8.6%) LA width, cm, mean SD LA length, cm, mean SD MV regurgitation, (n (%)) 162/183 (88.5%) 61/70 (87.1%) LA pressure, mm Hg, mean SD Procedure time, min, mean SD TIA indicates transient ischemic attack; MV, mitral valve. Evolving percutaneous therapies for various cardiac conditions are being developed, many of which require use of large-diameter transseptal sheaths. For example, novel balloon ablation strategies for atrial fibrillation with the endoscopic laser balloon 2 or cryoballoon catheters 8 require access to the LA with either a 14F or 15F outer diameter transseptal sheath, respectively. Likewise, various LAA closure devices use sheaths with outer diameters ranging from 12F to 15F. 1,9 Thus, knowledge of the natural history of iasds associated with large transseptal sheaths is vital, given the potential widespread clinical applicability of the above mentioned technology in patients with atrial fibrillation. As expected, the majority of patients in this series had an iasd detected immediately after the procedure. The only clinical factor that was significantly different between the group with and that without a postprocedural iasd was Table 3. the LA pressure at the time of device implantation. Not only was the LA pressure difference significantly higher in patients with an iasd, but this value was elevated above the upper limits of normal for LA pressure. Although not confirmed in a prior study, 10 these data suggest that LA pressure influences the patency and/or detection of an iasd on TEE imaging. Additionally, because most patients with atrial fibrillation have near normal right and left atrial pressures in the absence of significant valvular heart disease, the predominance of left-to-right shunting in patients with an iasd is suggestive of ongoing elevations in LA pressure. Although the clinical consequences (that is, paradoxical embolism) of right-to-left shunting in patients with an iasd is frequently of concern, one cannot ignore the possible detrimental consequences of left-toright cardiac shunting (right heart volume overload and pulmonary hypertension). Adverse consequences of leftto-right intracardiac shunting was not noted in this series of patients; this probably is related to the small size of the iasd and accompanying small shunt fraction, which is likely to have minimal clinical sequelae. In this study, there was no clinical evidence of right heart volume overload and pulmonary hypertension. Although the findings of our work suggest that the use of larger-diameter transseptal sheaths may not be associated with a significant increase in the incidence and complications related to the presence of an iasd, it must be recognized that the these findings are limited to a specific population of patients undergoing transseptal catheterization with placement of a 14F outer diameter sheath. It is possible that the incidence and complications of iasd may be higher when larger diameter transseptal sheaths are used. For example, Rilling et al 10 reported on 40 patients undergoing a single-puncture, double-transseptal procedure with 14F and 8.5F outer diameter sheaths. The prevalence of iasd was 21.1% at 6 months. Hammerstingl et al 5 reported on 27 patients who underwent a single transseptal puncture followed by the combined placement of both an 8F transseptal sheath and an adjacent catheter though the single puncture site. Approximately 30% of patients were noted to have a persistent iasd at 9-month follow-up. Thus, larger-diameter sheaths might result in a Stroke/Systemic Embolism During the Follow-Up Period in Patients With and Without iasd Events During the Follow-Up Period/Patient-Years of Follow Up (Rate Per 100 Patient-Years) Risk Factor/Time Point Shunt Present Shunt Absent Hazard Ratio (95% CI) P Value Postprocedure 5/438.7 (1.1) 4/155.7 (2.6) 0.45 ( ) 0.24 Postprocedure, adjusted for CHADS 2 score 0.46 ( ) Months 1/31.6 (3.2) 8/562.8 (1.4) 2.35 ( ) Months adjusted for CHADS 2 score 1.16 ( ) Months, only stroke/systemic embolism after 1/30.1 (3.3) 5/549.8 (0.9) 4.08 ( ) mo assessed 12 Months, only stroke/systemic embolism after 2.14 ( ) mo assessed adjusted for CHADS 2 score

5 170 Circ Arrhythm Electrophysiol April 2011 higher rate of iasd. Further work examining the incidence and clinical consequence of iasds with very largediameter transseptal sheaths such as the 21F sheath required for the percutaneous left ventricular assist device 11 and the 24F sheath required for the percutaneous mitral valve clip 12 is prudent. Limitations First, the study was a retrospective post hoc analysis of data derived from a randomized clinical trial. As such, immeasurable differences between the groups may be present; however, no large differences in the most important demographic, clinical, and procedural characteristics were noted (see Table 2). Second, the sensitivity of TEE for the detection of an iasd is not 100%. 13 However, this is a commonly used tool in clinical practice to detect iasd. Moreover, the concomitant use of saline contrast and, on occasion, color Doppler, allows for optimization of the sensitivity of this test for detection of clinically important iasd. Third, all patients in this study were on warfarin therapy for at least 45 days, followed by aspirin and clopidogrel therapy for 4.5 months thereafter. It is possible that the use of anticoagulant and antiplatelet agents may have reduced the risk of paradoxical embolization associated with an iasd. However, this is unlikely to have invalidated our primary findings as the risk or stroke/systemic embolism was also low during the follow-up period when anticoagulation was withdrawn. In addition, current clinical practice warrants that most patients undergoing left atrial procedures such as atrial fibrillation ablation are maintained on anticoagulant and/or antiplatelet agents for at least 3 to 6 months after the procedure. Fourth, it is possible that the higher than usual left atrial pressures ( mm Hg) encountered in this patient population minimized the incidence of right-to-left shunting noted on TEE imaging and hence the rate of paradoxical embolism. While possible, the clinical characteristics of patients in this cohort are typical of patients undergoing other left heart procedures with transseptal catheterization, and thus our results remain clinically relevant. Fifth, the overall clinical event rate was small, and clinical events simply reported as stroke or systemic embolism and not adjudicated as to whether they may be related to a paradoxical embolism. Thus, it is likely that the clinical event rate specifically related to paradoxical embolism secondary to the presence of an iasd is even lower. Sixth, other complications associated with iasds such as migraines were not reported. Given the low incidence of migraines reported in other series patients undergoing transseptal catheterization, 14 our primary clinical findings stand on their own. The strength of our work lies in the sample size, which is the largest to date reporting on the incidence of iasd in patients undergoing transseptal catheterization, the long follow-up period, systematic imaging protocol, and reporting of clinical events. In summary, patients undergoing transseptal catheterization procedures with a large-diameter transseptal sheath have a high spontaneous closure rate of iasds and no increased rate of stroke/systemic embolization during long-term follow-up. The consistency of these findings with previously reported work speaks to the safety of iasds associated with routine transseptal catheterization with 12F and smaller diameter sheaths. Systematic imaging to assess for iasd in patients who have undergone routine transseptal catheterization is not necessary, nor is the institution of specific therapy if this incidental finding is discovered. Sources of Funding Funding for the PROTECT AF study was provided by Atritech (Plymouth, MN). Disclosures Drs Reddy and Douglas have received clinical grant support from Atritech. References 1. Holmes DR, Reddy VY, Turi ZG, Doshi SK, Sievert H, Buchbinder M, Mullin CM, Sick P, for the PROTECT AF Investigators. Percutaneous closure of the left atrial appendage versus warfarin therapy for prevention of stroke in patients with atrial fibrillation: a non-inferiority randomized trial. Lancet. 2009;374: Reddy VY, Neuzil P, Themistoclakis S, Danik SB, Bonso A, Rossillo A, Raviele A, Schweikert R, Ernst S, Kuck KH, Natale A. Visually guided balloon catheter ablation of atrial fibrillation: experimental feasibility and first-in-human multicenter clinical outcome. Circulation. 2009;120: Schmidt B, Tilz R, Kars Neven K, Chun KR, Fuernkranz A, Ouyang F. Remote robotic navigation and electroanatomical mapping for ablation of atrial fibrillation: considerations for navigation and impact on procedural outcome. Circ Arrhythm Electrophysiol. 2009;2: Obel O, Mansour M, Picard M, Ruskin J, Keane D. Persistence of septal defects after transseptal puncture for pulmonary vein isolation procedures. Pacing Clin Electrophysiol. 2004;27: Hammerstingl C, Likfett L, Jeong KM, Troatz C, Wedekind JA, Tiemann K, Luderitz B, Lewalter T. Persistence of iatrogenic atrial septal defect after pulmonary vein isolation: an underestimated risk? Am Heart J. 2006;152:362.e1 362.e5. 6. Rilling A, Meyerfeldt U, Birkemeyer R, Treusch F, Kunze M, Jung W. Persistent atrial septal defect after pulmonary vein isolation: incidence and clinical implications. J Interv Card Electrophysiol. 2008;22: De Ponti R, Cappato R, Curnis A, Della Bella P, Padeletti L, Raviele A, Santini M, Salerno-Uriarte JA. Trans-septal catheterization in the electrophysiology laboratory: data from a multicenter survey spanning 12 years. J Am Coll Cardiol. 2006;47: Chierchia GB, de Asmundis C, Muller-Burri SA, Sarkozy A, Capulzini L, Paperella G, Chierchia S, Roos M, Brugada P. Early recovery of pulmonary vein conduction after cryoballoon ablation for paroxysmal atrial fibrillation: a prospective study. Europace. 2008; 11: AGA Medical Corporation. Amplatzer Cardiac Plug: Instructions for use. Available at: docs/ _amplatzer_cardiacplug_instructions_for_use_ OUS_Only_01_11_10.pdf. 10. Rilling A, Meyerfeldt U, Kunze M, Birkemeyer R, Miljak T, Jackle S, Hajredini B, Treusch F, Jung W. Persistent iatrogenic atrial septal defect after a single puncture, double transseptal approach for pulmonary vein isolation using a remote robotic navigation system: results from a prospective study. Europace. 2010;12: Sur JP, Pagani FD, Moscucci M. Percutaneous closure of an iatrogenic atrial septal defect. Catheter Cardiovasc Interv. 2009;73:

6 Singh et al iasd After Transseptal Puncture With a 12F Sheath Ussia GP, Barbanti M, Tamburino C. Feasibility of percutaneous transcatheter mitral valve repair with the MitraClip system using conscious sedation. Catheter Cardiovasc Interv. 2010;75: Kronzon I, Ruiz CE. Diagnosing patent foramen ovale: too much or too little. J Am Coll Cardiol Cardiovasc Imaging. 2010;3: Chilukuri K, Sinha S, Berger R, Marine JE, Cheng A, Nazarian S, Scherr D, Spragg D, Calkins H, Henrikson CA. Association of transseptal punctures with isolated migraine aura in patients undergoing catheter ablation of cardiac arrhythmias. J Cardiovasc Electrophysiol. 2009;20: CLINICAL PERSPECTIVE The increasing complexity of procedures involving the left atrium has resulted in the use of larger-diameter transseptal sheaths, which raises concern for persistent iatrogenic atrial septal defects (iasds) that may be associated with paradoxical embolic events. This study assessed the incidence of and clinical consequences of persistent iasds related to transseptal catheterization with a 12F transseptal sheath. Although the incidence of iasd is high immediately postprocedure, 10% of patients had a residual iasd at 1-year follow-up. In addition, no significant difference in embolic events was observed during the follow-up period in patients with and without a residual iasd. These findings suggest that transseptal catheterization with larger-diameter sheaths is not associated with additional long-term adverse consequence. Further work to assess the safety of very-large-diameter sheaths is prudent.