Author + information
- Received December 18, 2017
- Revision received February 26, 2018
- Accepted April 5, 2018
- Published online August 20, 2018.
- Jackson J. Liang, DOa,
- Sanghamitra Mohanty, MD, MSa,
- Joe Fahed, MDa,
- Daniele Muser, MDa,
- David F. Briceno, MDa,
- J. David Burkhardt, MDb,
- Jeffrey S. Arkles, MDa,
- Gregory E. Supple, MDa,
- David S. Frankel, MDa,
- Saman Nazarian, MD, PhDa,
- Fermin C. Garcia, MDa,
- David J. Callans, MDa,
- Sanjay Dixit, MDa,
- Luigi Di Biase, MD, PhDb,c,
- Andrea Natale, MDb,
- Francis E. Marchlinski, MDa and
- Pasquale Santangeli, MD, PhDa,∗ ()
- aDivision of Cardiology, Electrophysiology Section, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania
- bTexas Cardiac Arrhythmia Institute, Austin, Texas
- cMontefiore Medical Center, Albert Einstein College of Medicine, New York, New York
- ↵∗Address for correspondence:
Dr. Pasquale Santangeli, Division of Cardiology, Electrophysiology Section, Hospital of the University of Pennsylvania, 9 Founders Pavilion–Cardiology, 3400 Spruce Street, Philadelphia, Pennsylvania 19104.
Objectives This study reports outcomes of bailout atrial balloon septoplasty (ABS) to overcome challenging left atrial (LA) access in patients undergoing atrial fibrillation (AF) ablation.
Background Transseptal puncture (TSP) and LA access for AF ablation can be challenging in patients with prior atrial septal surgery, percutaneous closure, or scarred septum due to multiple prior TSPs.
Methods The study identified patients who underwent AF ablation at 2 ablation centers from 2011 to 2017 with challenging TSP in whom bailout percutaneous ABS was performed to allow LA access. Following TSP, the transseptal sheath could not be advanced to the LA despite multiple attempts or approaches including use of a stiff wire sequentially in the left and right pulmonary veins, use of a stiff pigtail exchange wire advanced in the LA or left ventricle, or sequential dilation with progressively larger diameter long dilators. ABS was performed using a noncompliant balloon (diameter 4 to 10 mm) advanced over a stiff wire deployed in the left superior pulmonary vein, allowing passage of the transseptal sheaths for completion of the AF ablation procedure.
Results Fifteen patients (mean age 54.4 ± 15.5 years, 9 women) with challenging TSP (7 patients with prior surgical ASD repair, 2 with percutaneous ASD closure devices, and 13 with ≥1 previous TSP) underwent bailout ABS for AF ablation. After TSP (radiofrequency assisted in 10 cases), ABS was successful and permitted access to the LA for ablation in all patients. Mean time required to perform ABS was 21.3 ± 19.4 min, and mean total procedure time was 241.1 ± 114.6 min (fluoroscopy time 62.0 ± 29.9 min). There were no procedural complications.
Conclusions In patients undergoing AF ablation with difficult transseptal access due to scarred, surgically, or percutaneously repaired atrial septum, ABS is a safe and effective bailout strategy to obtain transseptal access.
Catheter ablation is an effective and widely accepted treatment option for patients with atrial fibrillation (AF) refractory to antiarrhythmic drugs (1,2). Pulmonary vein isolation (PVI) is the cornerstone for AF ablation and access to the left atrium (LA) via transseptal puncture (TSP) is necessary to perform PVI. Patients who have undergone prior AF ablation may have septal scarring and thickening at the sites of prior transseptal access, complicating repeat TSP (3,4). Additionally, TSP may be difficult in patients who have previously undergone surgical or percutaneous repair for atrial septal defects (ASDs). Furthermore, certain ablation platforms such as cryoballoon ablation may require larger sheaths which can be challenging to advance through the septum. Isolated case reports have described the use of atrial balloon septoplasty (ABS) to dilate the access site and allow successful passage of the sheath to the LA (5–7). However, the efficacy and safety of ABS to obtain LA access in larger patient cohorts remains unknown. We report a series of patients undergoing catheter ablation of AF with challenging TSP in whom bailout ABS was performed to obtain LA access.
We identified all patients in whom ABS was performed to allow for transseptal access during AF ablation procedures at the Hospital of the University of Pennsylvania and the Texas Cardiac Arrhythmia Institute. Baseline clinical and demographic information and details of the AF ablation procedure were collected from patient medical records. All patients signed a written informed consent according to institutional guidelines at both centers, and data entered in the corresponding institutional registries were approved by the respective centers’ Investigational Review Boards.
AF ablation procedure
The standard mapping and AF ablation techniques at both centers have been described previously (8,9). All patients were brought to the electrophysiology lab in the fasting state and electrocardiographic and hemodynamic monitoring was initiated. Patients were prepped and draped in the usual sterile fashion. General anesthesia with intubation and mechanical ventilation was used in all cases, with the addition of high-frequency jet ventilation for the cases performed at the Hospital of the University of Pennsylvania. Femoral venous (or internal jugular venous) access was obtained via modified Seldinger technique under ultrasound guidance and radial arterial access was obtained for continuous blood pressure monitoring throughout the case. A diagnostic intracardiac echocardiography (ICE) catheter (8-F, AcuNav, Biosense Webster, Diamond Bar, California) was advanced into the right atrium via the left femoral vein. A decapolar catheter was placed along the posterior right atrium from the left femoral vein. The coronary sinus catheter type and access approach differed between the 2 centers (either a decapolar catheter from the left femoral vein or a 20-pole catheter from the right internal jugular vein). The technique for transseptal access including bailout ABS is described in detail subsequently. After transseptal access was obtained, electroanatomic mapping and radiofrequency ablation was performed using wide antral PVI together with ablation of spontaneous or inducible non-PV triggers (10,11). The techniques for induction and localization of non-PV triggers have been described in detail previously (11). In brief, all patients underwent infusion of high doses of isoproterenol (up to 20 to 30 μg/min) to induce non-PV triggers. Atrial burst pacing and cardioversion of spontaneous/inducible AF together with infusion of isoproterenol was also performed for non-PV trigger induction in cases done at the Hospital of the University of Pennsylvania (11). In 1 patient PVI was performed using a 28-mm Arctic Front Advance (Medtronic, Minneapolis, Minnesota) cryoballoon ablation catheter as previously described (12).
Transseptal access and ABS
Intravenous heparin was administered before the transseptal access to obtain a target activated clotting time of >300 s. Double (n = 13) or single (n = 2) TSP was performed from the right femoral vein to facilitate LA ablation. For cases utilizing radiofrequency ablation, 2 long 8.5-F transseptal sheaths (Agilis [8.5-F] and SL-1 [8.5-F] or SL-0 [8.5-F] and LAMP-90 [8.5-F], St. Jude Medical, St. Paul, Minnesota) were advanced over a long guidewire to the superior vena cava under fluoroscopic guidance. In 1 patient, single transseptal access was obtained using an 8.5-F Merit HeartSpan sheath (Merit Medical Systems, South Jordan, Utah). In the patient who underwent cryoballoon ablation, a FlexCath Advance steerable sheath (12-F inner diameter, 15-F outer diameter; Medtronic) was advanced over a stiff exchange wire positioned in the left superior PV (LSPV) once LA access was obtained with an SL-1 sheath using the technique described subsequently. A flushed BRK transseptal needle (St. Jude Medical) or a radiofrequency transseptal needle (NRG Transseptal Needle, Baylis Medical, Montreal, Canada) was introduced into the Agilis/SL-0/SL-1 sheaths, and the entire system was gently withdrawn under orthogonal fluoroscopic and ICE visualization. The optimal site of TSP access, as determined with ICE imaging, typically corresponded to the posteroinferior portion of the fossa ovalis with far-field visualization of the left PVs. A slightly more anterior access was performed in the patient who underwent cryoballoon ablation. Crossing of the interatrial septum with the needle was verified with ICE, fluoroscopy, and LA pressure recording; advancement of the transseptal dilator over the needle was possible in all cases. However, significant resistance was encountered when attempting to advance the transseptal sheath over the dilator and needle (Online Video 1). At this point the following strategies were sequentially attempted: a 0.035-inch stiff guidewire (Amplatz Super Stiff, Boston Scientific, Marlborough, Massachusetts; or Amplatz Extra-Stiff, Cook Medical Inc., Bloomington, Indiana) was advanced into the LSPV followed by the right superior PV, or a 0.025-inch stiff pigtail exchange guidewire (TorayGuide guidewire, Toray Group, Toray Industries, Inc., Tokyo, Japan; or Protrack, Baylis Medical) was advanced to the LA or left ventricle to provide more support to the dilator and facilitate the advancement of the sheath in the LA. In 4 patients, additional attempts at sequential dilation using progressively larger diameter (9-F, 11-F, then 14-F) long dilators advanced over a stiff guidewire positioned into the LSPV were also performed. If the sheath still could not pass through the septum, ABS was attempted (Online Video 1). A stiff guidewire (260-cm Amplatz Super Stiff, Amplatz Extra-Stiff, Protrack, or TorayGuide) was advanced to the LSPV (Amplatz wire) or LA (Protrack or TorayGuide wire), the dilator was pulled back, and the sheath was positioned in the mid right atrium. A noncompliant 0.035-inch POWERFLEX Pro PTA Dilation Catheter (diameter 4 to 10 mm, length 2 to 15 cm; Cordis, Milpitas, California) was advanced over the wire through the sheath and positioned across the septum as visualized with fluoroscopy and ICE. The balloon was fully or partially inflated under fluoroscopic and direct ICE visualization to assess for septostomy dilation. Once the septostomy site was appropriately dilated, the balloon was deflated and the sheath was advanced over the balloon and wire in the left atrium (Online Video 1). Figures 1 and 2⇓⇓ show sequential fluoroscopic and ICE images, respectively, demonstrating the ABS technique.
After ablation, patients were monitored in the hospital for at least 1 day and assessed for procedural complications. Transthoracic echocardiograms were not routinely performed for asymptomatic patients before hospital discharge. Standard post-ablation follow-up included provider visits at 6 weeks, 6 months, and 1 year post-ablation.
Between March 2012 and November 2017, of 8,682 patients undergoing AF ablation at our institutions, 15 (0.17%) patients underwent bailout ABS to obtain transseptal access (mean age 54.4 ± 15.5 years, 9 women). Ten (66.7%) of these patients underwent catheter ablation for nonparoxysmal forms of AF or atypical flutter. Mean LA diameter was 5.0 ± 0.7 cm and mean left ventricular ejection fraction was 59.7 ± 7.1%. Clinical and demographic characteristics for all patients are listed in Table 1. Seven patients had undergone prior open surgical ASD repair, in 2 of whom a GORE-TEX soft tissue patch (W.L. Gore & Associates, Inc., Newark, Delaware) was used. Two patients had undergone percutaneous ASD closure with Amplatzer devices. At least 1 prior procedure involving TSP had been previously performed in 13 patients (1 prior procedure in 3 patients, 2 procedures in 6 patients, and 3 procedures in 4 patients). In 2 patients, no prior cardiac surgery/closure device or TSP procedures had been performed (Patients #3 and #8) (Table 1): 1 of these patients had severe lipomatous infiltration of the interatrial septum and there was difficulty passing the larger cryoballoon transseptal sheath, whereas the other had evidence of severely aneurysmal interatrial septum.
Procedural details and bailout ABS outcomes
Procedural details and procedural outcomes are reported in Table 2. One patient underwent cryoballoon ablation while the remaining 14 underwent radiofrequency ablation. The mean total procedure time was 241.1 ± 114.6 min and the mean total fluoroscopy time was 62.0 ± 29.9 min.
The TSP was assisted by radiofrequency energy in 10 (66.7%) patients: in 1 patient standard electrocautery in the “cut” mode (Bovie Medical, Purchase, New York) was applied at 20 to 40 W at the back end of a BRK transseptal needle whereas in the remaining patients a radiofrequency transseptal needle (NRG Radiofrequency Transseptal Needle) was used to achieve passage of the needle into the LA. ABS was successful in all cases and allowed for passage of the sheath into the LA and completion of the ablation procedure. The mean duration to achieve transseptal access was 70.6 ± 32.1 min. Of this total duration, 49.5 ± 21.8 min were spent with multiple unsuccessful attempts to obtain access before ABS using the techniques described previously. The ABS procedure itself took a mean of 21.3 ± 19.4 min and was performed during the same procedure immediately after other strategies for transseptal access had failed. In all but 1 patient, ABS was performed over a stiff guidewire (Amplatz Super Stiff) positioned in the left superior PV. In 1 case, ABS was performed over a stiff pigtail guidewire (Protrack) deployed in the left atrium. A total of 1 to 6 balloon inflations at 4 to 14 atm were required to allow passage of the sheaths into the LA (Table 2). Two patients required 2 separate ABS (with the same balloon) for both sheaths to be advanced into the LA. One patient underwent single TSP followed by ABS with a 6-mm balloon and both sheaths were inserted through the same access using a retained wire technique. In the patient undergoing cryoballoon ablation, the 8.5-F SL-1 sheath was able to pass through the TSP site, but the larger, 12-F FlexCath could not pass through the lipomatous septum until ABS was performed. In the remaining patients, 1 of the 2 sheaths could be passed into the LA while the second sheath could not be advanced and required ABS to pass to the LA. In both patients with Amplatzer closure devices, ABS was performed in the native septum posterior and inferior to the devices.
Acute procedural outcomes and follow-up
There were no procedural complications. After a median 167.0 days (interquartile range: 7.5 to 360.5 days) follow-up, cumulative AF-free survival was 66.7% overall. Follow-up echocardiography was available for review in 14 (93%) patients and was performed a median of 114 days after ABS (Table 1). In 1 of the 2 patients in whom ABS was performed through a GORE-TEX soft tissue patch ASD repair, a positive bubble study consistent with ASD was present on follow-up echocardiogram 6 weeks post-ablation. This patient underwent a repeat AF ablation 18 months following the index procedure, and a residual ASD was confirmed on ICE; the LA access during the repeat procedure could be obtained through the residual ASD without need for repeat TSP or ABS. This patient remained asymptomatic during follow-up with regard to the ASD. No intervention was performed for the ASD and the patient was maintained on anticoagulation. In the 13 remaining patients with follow-up echocardiograms, no residual ASD was present (Table 2). In addition to the patient with the persistent ASD, 2 additional patients underwent repeat AF ablation during follow-up and ABS was again required to achieve transseptal access in both patients.
The present study reports on the largest series of AF patients in whom transseptal access could not be achieved with standard approaches, and required bailout ABS to obtain LA access. Prior descriptions of ABS to obtain LA catheterization have been limited to few single-patient case reports (5–7). The results of our study document the high efficacy and safety profile of ABS. In particular (5–7), ABS was successful and permitted passage of the transseptal sheath to allow AF ablation in all patients, and there were no complications associated with the procedure. These findings support the inclusion of ABS to the electrophysiology procedural armamentarium as a technique to achieve LA access in patients with challenging transseptal access.
As the number of patients undergoing AF ablation continues to increase, a significant number of patients referred for ablation may have undergone prior atrial septal repair surgeries or previous AF ablation attempts in which TSP was performed. In these patients, septal scarring may complicate TSP, making it difficult to access the LA. Several tools and strategies to achieve transseptal access have been proposed, including the delivery of radiofrequency energy from the needle tip and the use of the SafeSept nitinol guidewire (Pressure Products, Santa Barbara, California) (13–18). Although these strategies are effective for allowing successful passage of the transseptal needle across the septum into the LA, in cases where the atrial septum is severely scarred, it may be impossible to pass the larger-diameter outer sheath across. Other strategies include the use of stiff wires advanced into the LSPV or right superior PV or the use of stiff pigtail wires to provide more support for the dilator and sheath. We have also previously described the technique of pre-dilating the septostomy over a stiff wire with progressively larger diameter (up to 14-F) long dilators to facilitate passage of a transseptal sheath in patients in whom TSP was performed through ASD closure devices (19). In the present series, pre-dilation of the septostomy site with long dilators of progressively larger size was attempted without success in 4 cases.
In our experience, it is more difficult to obtain transseptal access with stiffer sheaths (such as the Agilis) compared with sheaths with lower outer profiles (e.g., the standard 8.5-F SL sheaths). This is likely due to the presence of a substantial diameter mismatch between the dilator profile and the outer diameter of the Agilis sheath due to the need to accommodate the steering mechanism within the body of the sheath. As a result, there is a significant outer diameter “step-up” between the dilator and the Agilis sheath, which makes advancement of the sheath through the interatrial septum over the dilator more challenging compared with standard nonsteerable transseptal sheaths.
We have previously shown that TSP and LA access for AF ablation can be safely performed in patients with prior atrial septal repair or multiple previous TSP attempts (3). In that prior experience, while TSP was straightforward in the majority of cases using standard approaches, “challenging” TSP (requiring stiff wires or necessitating multiple attempts) were encountered in 5% of patients with multiple (≥2) prior LA ablation procedures and 26% patients with prior atrial septal repair. Of note, TSP failure was reported in 1% in the multiple TSP group and 4% in prior atrial septal repair group (3), and ABS was not attempted in any of those patients. The present study expands on our prior experience by showing the high efficacy of ABS to obtain transseptal access when standard approaches fail. Importantly, the described ABS technique has been utilized only as a “bailout” strategy in the rare situation that all other techniques failed to permit transseptal access. Specifically, ABS has only been required for transseptal access in 0.17% of patients undergoing AF ablation at our 2 centers since we started using the technique. Before the introduction of ABS, there were 5 patients between the 2 study centers in whom transseptal access failed (2 with multiple prior TSPs, and 3 with ASD patch repairs) and LA access could not be achieved. Since we began using the ABS technique, there have been no patients at either center in whom transseptal access could not be achieved. Upon transseptal access following ABS, we could not detect any qualitative difference in catheter maneuverability or torqueability within the LA compared versus standard cases.
The main concern with the use of ABS is the creation of persistent ASD. To minimize the occurrence of this complication, careful attention was paid to size the balloon appropriately (i.e., slightly larger than the transseptal sheath outer diameter) and in some cases, only low-pressure partial balloon inflation was required to dilate the septum enough to allow passage of the transseptal sheaths. As such, the size of the ASD created with our ABS technique should be comparable to what is routinely created with standard transseptal sheaths. The occurrence of persistent ASD following transseptal access with large sheaths in patients undergoing catheter ablation for AF is well known. For example, in 1 study, 22% of patients treated with cryoballoon ablation with a 15-F transseptal sheath had persistent ASD compared with 8.5% of those treated with double transseptal access using two 8-F sheaths with radiofrequency catheter ablation after 11.6 months’ follow-up (20). In another series of 42 patients undergoing TSP for AF ablation, Hammerstingl et al. (21) reported that the risk of persistent ASD is higher in patients treated with a single TSP through which an ablation catheter plus an 8-F sheath are advanced into the LA (8 of 27 [29.6%] developed ASD) as opposed to 2 separate transseptal sheaths (0 of 15 [0%]). Typically, when iatrogenic ASDs occur post-TSP, they frequently will resolve spontaneously over time. Rillig et al. (22) examined rates of ASD after AF ablation via a single puncture, double-transseptal approach with remote robotic navigation system using both an 8.5-F SL-0 sheath and a 14-F sheath for the robotic catheter. Of the 40 total patients in their series, ASD was detected in 95% on day 1 post-ablation, 48% at 3 months, and 20% at 6 months. They found that of the 38 patients with an ASD on day 1, 30 (79%) had closed spontaneously by 6 months. Singh et al. (23) showed similar results in a subset of 253 patients from the PROTECT-AF (Percutaneous closure of the left atrial appendage versus warfarin therapy for prevention of stroke in patients with atrial fibrillation) study in whom a 12-F transseptal sheath was used to deploy LA appendage occlusion devices. In their study, ASD was seen in 87% immediately post-procedure, 34% at 45 days, 11% at 6 months, and only 7% at 1 year.
In our study, 14 (93%) patients underwent a follow-up echocardiographic study to assess for residual ASD, which was seen in just 1 (7%) patient. Of note, we found no evidence of residual ASD in all patients in which ABS had been performed through the native septum or prior pericardial patch repair, which is consistent with the observations of prior studies (6,7). However, a persistent ASD was detected in 1 of the 2 patients in whom ABS had been performed through a GORE-TEX patch.
Although no definite conclusions can be made based on the small sample size of our study, it is possible that the likelihood of persistent interatrial shunt after septoplasty may differ when ABS is performed in the native septum versus prosthetic patches. Although the native atrial septum may be capable of healing after ABS resulting in spontaneous closure, ABS through a synthetic patch may be less likely to heal and thus more likely to result in permanent ASD.
This was a retrospective study with a small sample size and a short mean follow-up duration. As per our standard of care for AF ablation, residual ASD was qualitatively assessed with color Doppler at ICE immediately post-procedure and it did not appear to be bigger than what is usually observed for standard AF cases where ABS is not used. This finding is likely due to a careful choice of balloon diameters comparable to the outer diameters of standard transseptal sheaths. However, a formal measurement of the size of the residual ASD as visualized by post-procedure ICE was not performed, and this is a potential limitation. Post-procedure follow-up echocardiography was done with transthoracic echocardiography and we do not routinely perform transesophageal echocardiography to assess for persistent ASD in asymptomatic patients after ABS. We acknowledge that transthoracic echocardiography may be less sensitive to detect ASD than transesophageal echocardiography. One patient transitioned their care elsewhere after their procedure and we were unable to contact this patient for repeat imaging. As such, we are unable to confirm the true incidence of iatrogenic ASD after ABS. Finally, pre-dilation of the septostomy with dilators of increasing size was attempted only in 4 patients because larger-diameter long dilators were not readily available in the electrophysiology lab for the remaining cases. As such, it is possible that pre-dilation with larger-diameters dilators might have been successful in some of the cases that required ABS in this study.
In AF patients with scarred or thickened LA septum undergoing catheter ablation, bailout ABS can be safely performed and is an effective adjunctive strategy to obtain transseptal access.
COMPETENCY IN MEDICAL KNOWLEDGE: In patients with scarred, surgical, or percutaneously repaired atrial septum, TSP and LA access can be difficult. ABS is a safe and effective bailout strategy to facilitate passage of the transseptal sheath into the LA after other strategies had failed.
TRANSLATIONAL OUTLOOK: Larger series are necessary to confirm the safety and efficacy of ABS before it can be used as mainstream strategy to facilitate transseptal access. Studies with longer follow-up duration and serial echocardiograms are necessary to determine the true incidence and significance of persistent ASD after ABS.
Dr. Nazarian has served as a consultant for Biosense Webster, CardioSolv, and Siemens; and has received research grant support from Biosense Webster. Dr. Di Biase has served as a consultant for Biosense Webster, Stereotaxis, Boston Scientific, and Abbott; and has received speaker/travel honoraria from Medtronic, Pfizer, and Biotronik. Dr. Natale has received consulting fees/honoraria from Biosense Webster, Inc., St. Jude Medical, Medtronic, and Boston Scientific. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- atrial balloon septoplasty
- atrial fibrillation
- atrial septal defect
- intracardiac echocardiography
- left atrium/atrial
- left superior pulmonary vein
- pulmonary vein isolation
- transseptal puncture
- Received December 18, 2017.
- Revision received February 26, 2018.
- Accepted April 5, 2018.
- 2018 American College of Cardiology Foundation
- January C.T.,
- Wann L.S.,
- Alpert J.S.,
- et al.
- Calkins H.,
- Kuck K.H.,
- Cappato R.,
- et al.
- Arkles J.,
- Zado E.,
- Supple G.,
- et al.
- Liang J.J.,
- Elafros M.A.,
- Muser D.,
- et al.
- Dixit S.,
- Lin D.,
- Frankel D.S.,
- Marchlinski F.E.
- Santangeli P.,
- Zado E.S.,
- Hutchinson M.D.,
- et al.
- Santangeli P.,
- Marchlinski F.E.
- Singh S.M.,
- Douglas P.S.,
- Reddy V.Y.