Author + information
- Received June 22, 2016
- Revision received September 12, 2016
- Accepted September 15, 2016
- Published online June 19, 2017.
- Ayman A. Hussein, MDa,∗ (, )
- Tanmay S. Panchabhai, MDb,
- Marie M. Budev, DOc,
- Khaldoun Tarakji, MDa,
- Amr F. Barakat, MDd,
- Walid Saliba, MDa,
- Bruce Lindsay, MDa and
- Oussama M. Wazni, MDa
- aSection of Cardiac Pacing and Electrophysiology, Cleveland Clinic, Cleveland, Ohio
- bJohn and Doris Norton Thoracic Institute, St. Joseph’s Hospital and Medical Center, Phoenix, Arizona
- cLung Transplant Program, Cleveland Clinic, Cleveland, Ohio
- dDepartment of Internal Medicine, Cleveland Clinic, Cleveland, Ohio
- ↵∗Address for correspondence:
Dr. Ayman A. Hussein, Cardiac Pacing and Electrophysiology, Department of Cardiovascular Medicine/J2-2, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, Ohio 44195.
Objectives The authors report their experience with atrial fibrillation (AF) rates and ablation findings in lung transplant recipients.
Background Pulmonary venous (PV) conduction recovery accounts for most failed atrial fibrillation (AF) catheter ablation procedures. Lung transplantation involves full surgical resection and replacement of the recipient's PVs with donor's PVs, which may represent the ultimate PV ablation.
Methods They followed 755 consecutive lung transplant recipients categorized based on transplant status (unilateral vs. bilateral) and pre-transplant AF.
Results In patients without pre-transplant AF (n = 704), late AF (beyond 6 months after transplant) occurred in 2.5% and 3.3% of unilateral or bilateral lung transplants, respectively. In patients with pre-transplant AF (n = 51), AF recurred in 19.4% and 25.0% of bilateral and unilateral transplants, respectively. In a subset of patients who underwent left atrial ablations after transplant for recurrent refractory AF (n = 8), PV conduction recovery across the surgical anastomoses lines was observed in 22 of 26 previously disconnected PVs. Conduction recovery was observed in ≥1 vein in all but 1 patient. Re-isolation of the veins with additional substrate modification/flutter ablations successfully restored and maintained sinus rhythm in 7 of 8 patients.
Conclusions In lung transplant recipients who undergo full surgical resection of the PVs, a prior history of AF was associated with late AF, regardless of whether patients underwent single or bilateral lung transplantation. PV conduction recovery still occurred and was observed in most patients who underwent left atrial ablation procedures for recurrent AF.
The role of the pulmonary veins (PVs) as the predominant source of triggers in atrial fibrillation (AF) is well-known (1–4). Catheter-based interventions to isolate the PVs have proven to be effective treatment strategies for AF. However, arrhythmias may still recur and are related mostly to pulmonary venous conduction recovery (3,5). It is generally thought that conduction recovery across ablation lines is primarily due to gaps between the ablation lesions or suboptimal lesion depth, which prevents transmural ablation (3).
Lung transplantation provides a unique opportunity to assess the efficacy of effective isolation of the veins as treatment for AF. In one of the largest cohorts reported in the literature, late AF, occurring >6 months after lung transplantation, was found in 0.5% and 12.6% of patients who had undergone bilateral and unilateral lung transplants, respectively, versus 11% in a thoracic surgery control group (6). However, late AF rates were still unknown in lung transplant patients who had pre-operative AF.
In a lung transplantation procedure, the PVs of the recipient are resected fully and replaced with the PVs and atrial cuff of a donor, a procedure that may represent the ultimate PV surgical ablation. The lung transplant population affords researchers the opportunity to assess success rates of a PV-only ablation, and further to assess whether PV conduction recovery is possible in this setting. This may offer greater insight into failed catheter-based ablations and their PV conduction recovery rates. We report our experience with AF rates and ablation findings in lung transplant recipients.
This study was approved by the Institutional Review Board at the Cleveland Clinic. All consecutive patients undergoing lung transplantation at our institution between 2006 and 2013 (n = 775) were enrolled in a prospectively maintained data registry. Patients with history of AF before lung transplant were identified by medical record review. The population was categorized into 4 groups at baseline, based on their history of pre-operative AF and whether they underwent single or double lung transplantation. The rates of late AF, which was defined as AF occurring more than 6 months after lung transplantation, were compared between the study groups. Electrical documentation of AF with an electrocardiogram, Holter monitor, or event monitoring was required for the diagnosis of AF.
In addition, all patients who underwent catheter ablation for AF at our institution between January 2000 and December 2015 were included in a prospectively maintained AF ablation registry. All consecutive patients with lung transplantation who had undergone subsequent AF ablation procedures were identified. The electrophysiological findings and outcomes of the ablation procedures are reported in this subset of patients (n = 8) with recurrent, drug-refractory AF who had undergone left atrial ablation procedures after lung transplantation.
Per institutional protocol, patients completed routine follow-up and close monitoring at the Cleveland Clinic post-transplant. Clinical care documentation was recorded via electronic medical records. For this study, we carried out an extensive review of medical records to assess the clinical characteristics of these patients and to record the incidence of late AF. This involved review of all available clinical documentation from office visits, inpatient notes, and discharge summaries. In addition, all correspondence between patients, their local treating physicians, and our clinical centers was documented via electronic records. Any clinical documentation from office visits, emergency department visits, or hospital admissions at other facilities was obtained and scanned into the electronic records.
Radiofrequency ablation and outcomes
Our AF ablation protocol and peri-procedural anticoagulation strategies have been previously reported in detail elsewhere (7–9). A phased array intravascular ultrasound catheter (Siemens AG Inc., Malvern, Pennsylvania) was placed in the right atrium to assist with trans-septal punctures, to guide catheter location and manipulation within the left atrium (LA), and to monitor for complications during the procedure. Multiple modalities were used to assess the anatomy of the LA to identify and delineate the lines of the surgical anastomoses, including 3-dimensional anatomic maps, voltage maps, pulmonary venograms, and real time intracardiac echocardiography. Catheter contact with tissue was confirmed with intracardiac echocardiography, fluoroscopy, and, more recently, contact-force–sensing catheters.
A circular mapping catheter (Lasso) was positioned at the ostia of each PVs to assess for PV potentials. Care was taken to ensure PV signal recording beyond the lines of surgical anastomoses, which allowed assessment of conduction recovery, dissociated firing, or entrance block with silent PVs. In the latter scenario, the Lasso was used to pace at the PV ostia and assess for bidirectional block. Demonstration of local capture was required in PV pacing maneuvers.
Radiofrequency ablation was performed to isolate the PVs that showed evidence of conduction recovery. The ablation strategy included ablation along the surgical anastomoses lines with extension of the ablation to the posterior wall and septal to the right-sided PVs at the discretion of the operators. In patients with single lung transplantation, the contralateral veins were also ablated.
After isolation of the PVs, programmed electrical stimulation was performed. For patients in whom atrial flutters (AFLs) were induced, activation and entrainment mapping were performed to identify the critical circuit of the flutter. Subsequent AFL ablation was performed as appropriate. Additional linear ablations were performed at the discretion of the operators. In patients with documented typical AFL, cavotricuspid isthmus ablation was performed with confirmation of bidirectional block by differential pacing.
After the ablation procedures, patients had clinical follow-up to assess for arrhythmia recurrence. Our follow-up and monitoring strategies post-AF ablation have been reported in detail elsewhere (5). Arrhythmia recurrence was defined as electrocardiographic documentation of an atrial tachyarrhythmia of at least 30 s duration, with or without symptoms, recorded on a 12-lead electrocardiogram, event recording, or Holter monitor.
Statistical analyses were performed by using the statistical software JMP Pro version 10.0 (SAS Institute, Cary, North Carolina). Continuous variables are reported as mean ± SD. Categorical variables are reported as number (percentage). The Student t test was used to compare means; the chi-square test was used for comparison of proportions, as appropriate. A 2-sided p < 0.05 was considered significant.
Patient population and baseline characteristics
A total of 775 consecutive patients underwent lung transplantation at our center between 2006 and 2013. Of these, 20 patients underwent heart and lung transplantation, and were excluded from the current study. The final study population therefore included 755 patients. The indications for lung transplantation were idiopathic pulmonary fibrosis (50.3%), chronic obstructive pulmonary disease (24.9%), cystic fibrosis (10.9%), idiopathic pulmonary arterial hypertension (5.3%), or other indications (8.6%).
In this population, 51 patients (6.8%) had prior history of AF; the remaining 704 patients (93.25%) had no prior history of AF (Figure 1). Of the 51 patients with prior AF, 31 underwent bilateral lung transplantation (group 1) and 20 underwent single lung transplantation (group 2). Of the 704 patients without prior AF, 427 underwent bilateral lung transplantation (group 3), and 277 underwent single lung transplantation (group 4). The baseline clinical characteristics of these patients are summarized in Table 1. There were no differences in the indications for lung transplantation among the groups (p = 0.20).
AF during follow-up
Early post-operative AF occurred in 83.9%, 80.0%, 43.1%, and 41.9% in groups 1 through 4, respectively (p across groups <0.0001). Over the course of follow-up (median: 779 days; 25th to 75th percentile: 317 to 1,454 days), 32 patients (4.2%) developed late AF, with AF occurrence in 19.4%, 25.0%, 3.3%, and 2.5% of patients in groups 1 to 4, respectively (p across groups <0.0001). In patients who underwent single lung transplantation (n = 297), late AF was observed during follow-up in 12 patients with no difference in AF incidence based on whether patients had right- or left-sided lung transplantation (p = 0.80).
Compared with patients who had no late AF, patients with late AF were more likely to have had pre-transplant AF (34.4% vs. 5.5%; p < 0.0001) and were also more likely to develop early post-operative AF (100% vs. 42.8%; p < 0.0001) (Table 2). Patients who developed late AF were also more likely to have a history of smoking (81.3% vs. 61.0%; p = 0.02). The immunosuppressive regimens were comparable between those with or without late AF, with the exception of mycophenolate mofetil, which was received by 50% of those with late AF (vs. 71.0% of those without late AF; p = 0.01). There were otherwise no differences in baseline clinical characteristics between patients with or without late AF.
Subset population with AF ablation procedures
Of the 32 patients who developed late AF after transplant, 6 underwent subsequent left atrial ablation procedures for recurrent symptomatic drug-refractory AF. From our AF ablation data registry, we identified 2 additional patients who underwent lung transplantation at other institutions and subsequently underwent left atrial ablation procedures for AF at our institution. The operative reports of the transplantations were obtained for both of these patients, and the final number of this subset of patients who underwent AF ablation procedures at our institution included 8 patients in total. Five of them had undergone bilateral lung transplant; 3 had undergone unilateral lung transplant. The transplant procedures included the surgical disconnection of 13 sets of ipsilateral veins, as well as anastomosis of the donor’s atrial cuff to the recipient’s atrium.
All 8 patients had post-operative AF within the week after transplant and all continued to have long-term recurrences beyond 6 months after surgery, qualifying as such as late AF. Only 1 patient had a history of AF before transplantation surgery and this was a recipient of bilateral lung transplant. The median time from transplant to ablation in this population was 43 months (25th to 75th percentile: 20 to 54 months).
Mapping and ablation
In all 8 patients, it was possible to determine the surgical anastomoses lines by multimodality assessment with evidence of scar, double potentials, or fractionated potentials by voltage mapping (Figure 2A) with additional direct visualization of atrial cuffs with pulmonary venography (Figure 2B), 3-dimensional maps (Figure 2C), or intracardiac echocardiography imaging. Of 26 surgically disconnected PVs, 22 veins had evidence of conduction recovery. The remaining 4 veins had bidirectional block. There was conduction recovery in at least 1 PV in 7 patients. In 1 patient who underwent a single left lung transplant, there was bidirectional block at the previously isolated veins.
In 18 of 22 veins with conduction recovery, there was evidence of LA-to-PV conduction, with PV potentials recorded at the veins’ ostia beyond the surgical anastomoses lines. The surgical lines encircling the atrial cuffs in these patients had normal electrical signals in multiple areas along the surgical anastomoses. Four veins (2 sets of ipsilateral veins in 2 patients) were silent, with evidence of entrance block but without exit block as assessed with pacing via the circular mapping catheter positioned at the PV ostia. Mapping of the surgical lines in these 2 patients showed scarring at the PV cuff LA anastomosis for both sets of PVs with areas of fractionated potentials. All 22 veins were isolated successfully, with ablation along the surgical anastomoses lines (Figure 3). In 4 patients, ablations were extended to the posterior wall and septal to the right-sided PVs.
After confirmation of isolation of the veins, 5 patients had left AFLs. Mapping was consistent with macro–re-entry using the scar at or around the surgical anastomoses lines (with entrance block to the veins) in 3 patients. These AFLs were isolated successfully with ablation near the anastomoses lines. Additional ablation was performed in the recipients’ atria to reinforce the surgical anastomoses lines. One patient had a perimitral flutter, which was ablated successfully with a right superior PV-to-mitral annulus line. The remaining patient had a roof-dependent flutter, which was ablated successfully at the roof with completion of a roof line to connect the contralateral atrial cuffs. In 1 patient who had undergone isolation of the PVs with posterior wall, roof, and gutter ablations, attempts to induce AFL resulted in AF and this was cardioverted to sinus rhythm.
Six patients underwent electrical isolation of the posterior wall with either extensive point-by-point ablation (n = 4) or with lines at the roof and floor of the atrium, which connected the ablation lines around the contralateral atrial cuffs (n = 2).
Post-ablation follow-up and outcomes
Over a median follow-up of 3 years after ablation, 6 of 8 patients (75%) remained arrhythmia free. Two patients had AF recurrence, 1 of whom underwent a redo ablation procedure. This procedure revealed limited focal conduction recovery at the surgical anastomosis line. Limited ablation was performed at this site, with successful isolation of the veins and no recurrence over 1 year of subsequent follow-up. There were no recurrences as AFL in any of the patients.
In this cohort, we report on pulmonary venous electrical conduction recovery and AF rates after full surgical resection and anastomosis of the PVs in lung transplant recipients who underwent surgical resection and anastomosis of atrial cuffs including the PVs as part of lung transplant procedures. In the overall cohort, which included 755 lung transplant recipients, late AF occurred in 4.2% of patients. A pre-operative history of AF was the main factor associated with late AF after transplantation. In patients with pre-operative AF, arrhythmia recurrence rates were 19.4% and 25.0% in bilateral and unilateral transplant recipients, respectively. In patients with no pre-operative history of AF, the risk of late AF was very low regardless of whether patients underwent unilateral or bilateral lung transplantation. An interesting finding in patients who underwent left atrial ablations after transplantation was that PV conduction recovery across the surgical anastomoses lines was observed in most of the previously disconnected PVs. Conduction recovery was observed in at least 1 vein in almost all patients who underwent left atrial ablations. Catheter ablation procedures in patients with recurrent AF, targeting re-isolation of the veins with additional substrate modification, successfully restored and maintained normal sinus rhythm.
Isolation of the PVs (1) remains the cornerstone of AF ablation procedures. In clinical practice, pulmonary venous conduction recovery is observed in most patients who undergo redo ablation procedures for recurrent AF (3,5), which seemed to have occurred in a significant proportion of patients with recurrent AF and left atrial ablations in our study.
In 1 series of patients with lung transplant and no pre-transplant atrial arrhythmias (10), the highest incidence of new-onset atrial arrhythmias occurred within 30 days after the transplant procedures and more commonly in patients with bilateral lung transplantation, which could be related possibly to perioperative inflammation. The same study assessed arrhythmia mechanisms in a subset of patients who underwent electrophysiologic testing and ablation and found multiple arrhythmia mechanisms in the majority of patients including peritricuspid flutter, perimitral flutter, right atrial incisional reentry, focal tachycardia from recipient’s PV antrum, focal PV fibrillation, and left atrial roof flutter (10). The arrhythmogenic foci-generating PV antral arrhythmias after transplant (including focal and reentrant) were localized to the recipient’s antrum at the level of the anastomosis lines with no evidence of connection across the surgical lines. In the literature, however, multiple studies have reported electrical conduction recovery across atrial surgical anastomoses lines after surgical disconnection of the veins, such as in patients who had undergone cardiac transplantation (11) or the “cut and sew” Cox Maze procedure (12). In patients who underwent lung transplantation, atrial tachycardia originating from the donor PV with conduction across surgical anastomosis to the recipient’s atrium has been also described (13). The mechanisms underlying electrical conduction across surgical scars remain unknown, but could be attributed to local cell-to-cell coupling (14).
Lung transplantation procedures provide a unique model for a surgical PV-only ablation strategy as treatment for AF. This model is different than orthotopic heart transplantation, which typically involves surgical disconnection of the recipient’s posterior wall of both atria (including the vena cavae) (15), or the “cut and sew” Cox maze procedure, which includes full surgical resection of the posterior LA in addition to multiple biatrial incisions (16).
Bilateral lung transplant recipients have been reported to be more likely to develop AF compared to heart transplant recipients despite full surgical isolation of the PVs in both procedures, suggesting that there are additional substrate-related factors which contribute to AF pathogenesis (17).
In a previously published report, late AF was significantly more common in single versus bilateral lung transplant recipients (6), which was not observed in our cohort, suggesting an additional role of the substrate in AF. Theoretically, PV conduction recovery would have occurred at similar rates across the study groups. Nonetheless, in bilateral lung transplant recipients who underwent surgical resection and implantation of all 4 PVs from donors, those with prior history of AF were much more likely to have recurrent late AF than those without pre-transplant AF. This suggests an important contribution of the substrate in patients with a history of AF. One important caveat in the interpretation of our findings is that PV electrical conduction recovery rate in the subset of patients with recurrent refractory AF and left atrial ablations does not necessarily reflect the PV recovery rate in the overall population of transplant recipients. Pulmonary venous conduction recovery could have occurred in patients without late AF without necessarily resulting in arrhythmogenesis. Another caveat is that no specific substrate assessment was performed, such as magnetic resonance imaging, rotor mapping, or complex fractionated electrogram mapping. Also, the role and contribution of the right atrium to arrhythmogenesis was not addressed specifically.
The atrial substrate plays an important role in AF pathogenesis. In patients with AF, the arrhythmia causes its self-perpetuation through progressive atrial remodeling (18–20). This involves multiple pathways, including inflammation, hemodynamic alterations, fibrosis, and atrial dilatation (20), which seem to be progressive over time (21). Another interesting observation in the current report was the decreased rate of AF in patients who received mycophenolate mofetil as part of their immunosuppressive regimen. This may be attributable to the anti-fibrotic properties of mycophenolate mofetil and, as such, its effect on the arrhythmogenic substrate (22).
The study has the inherent limitations of observational studies including residual confounding, but the data are largely derived from prospectively maintained databases. The study used medical records and may have missed asymptomatic AF, but this limitation would have affected the study groups to a similar extent and does not negate the observations regarding occurrence of late AF. In the subgroup of patients who underwent ablation procedures, the arrhythmogenic role of the PVs or the substrate were not systematically assessed and this subgroup was limited to a small number of patients. Furthermore, the catheter ablation procedures were not standard and uniform. Finally, care was taken to identify the surgical anastomoses lines in all patients undergoing ablation, but no standard protocol was used for that purpose.
In patients who had undergone surgical resection and anastomosis of the PVs as part of lung transplantation, a prior history of AF was associated with late AF, regardless of whether patients underwent single or bilateral lung transplant. In patients with pre-transplant AF, the transplant procedures eliminated AF in 75% to 80% of patients, but pulmonary venous conduction recovery still occurred and was observed in most patients who underwent left atrial ablation procedures for recurrent AF.
COMPETENCY IN MEDICAL KNOWLEDGE: In patients who had undergone surgical resection and anastomosis of the PVs as part of lung transplantation, a prior history of AF was associated with late AF, regardless of whether patients underwent single or bilateral lung transplant. The findings suggest important contribution of the atrial substrate to AF recurrences. In patients with pre-transplant AF, the transplant procedures eliminated AF in 75% to 80% of patients, but pulmonary venous conduction recovery still occurred and was observed in most patients who underwent left atrial ablation procedures for recurrent AF. Ablation procedures targeting re-isolation of the PVs with additional substrate modification were associated with low AF recurrence rate.
TRANSLATIONAL OUTLOOK: The mechanisms underlying electrical conduction recovery across surgical lines deserves further investigation. In patients with recurrent AF after full surgical resection and anastomosis of the PVs, characterization of the atrial substrate with electroanatomic mapping, rotor mapping, or magnetic resonance imaging may help better explain arrhythmogenesis in this setting.
The 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 fibrillation
- atrial flutters
- left atrium
- pulmonary venous/vein
- Received June 22, 2016.
- Revision received September 12, 2016.
- Accepted September 15, 2016.
- Chen S.A.,
- Hsieh M.H.,
- Tai C.T.,
- et al.
- Calkins H.,
- Kuck K.H.,
- Cappato R.,
- et al.
- January C.T.,
- Wann L.S.,
- Alpert J.S.,
- et al.
- Hussein A.A.,
- Saliba W.I.,
- Martin D.O.,
- et al.
- Bassiouny M.,
- Saliba W.,
- Rickard J.,
- et al.
- Chaikriangkrai K.,
- Jyothula S.,
- Jhun H.Y.,
- et al.
- Saoudi N.,
- Redonnet M.,
- Anselme F.,
- Poty H.,
- Cribier A.
- Wazni O.M.,
- Saliba W.,
- Fahmy T.,
- et al.
- See V.Y.,
- Roberts-Thomson K.C.,
- Stevenson W.G.,
- Camp P.C.,
- Koplan B.A.
- Kleber A.G.
- Vaseghi M.,
- Boyle N.G.,
- Kedia R.,
- et al.
- Dizon J.M.,
- Chen K.,
- Bacchetta M.,
- et al.
- Morillo C.A.,
- Klein G.J.,
- Jones D.L.,
- Guiraudon C.M.
- Wijffels M.C.,
- Kirchhof C.J.,
- Dorland R.,
- Allessie M.A.
- Hussein A.A.,
- Saliba W.I.,
- Barakat A.,
- et al.
- Roos N.,
- Poulalhon N.,
- Farge D.,
- Madelaine I.,
- Mauviel A.,
- Verrecchia F.