Author + information
- Received February 5, 2018
- Revision received April 2, 2018
- Accepted May 8, 2018
- Published online August 20, 2018.
- Katharina Schoene, MDa,b,∗ (, )
- Arash Arya, MDa,
- Cosima Jahnke, MDa,
- Ingo Paetsch, MDa,
- Sotirios Nedios, MDa,
- Sebastian Hilbert, MDa,
- Andreas Bollmann, MD, PhDa,
- Gerhard Hindricks, MDa,b and
- Philipp Sommer, MDa
- aDepartment of Electrophysiology, Heart Center, University of Leipzig, Leipzig, Germany
- bLeipzig Heart Institute, Leipzig, Germany
- ↵∗Address for correspondence:
Dr. Katharina Schoene, University of Leipzig, Heart Center, Department of Electrophysiology, Struempellstrasse 39, 04289 Leipzig, Germany.
Objectives The aim of the present study was to analyze and report a single-center experience with catheter interventional treatment of radiofrequency-induced pulmonary vein stenosis (PVS) following atrial fibrillation (AF) ablation.
Background Catheter interventional treatment of radiofrequency-induced PVS following AF ablation remains a challenging field because of a lack of randomized data and treatment guidelines.
Methods All patients at a single center who underwent catheter interventional treatment for radiofrequency-induced PVS were retrospectively assessed.
Results From January 2004 to September 2017, the total rate of PVS following interventional AF ablation was 0.78% (87 of 11,103). Thirty-nine patients with PVS were treated with 84 catheter interventions: 68 (81%) with percutaneous transluminal balloon angioplasty (PTA) and 16 (19%) with stent implantation. The distribution of stent type was 3 drug-eluting stents (19%) and 13 bare-metal stents (81%). The overall restenosis rate was 53% after PTA versus 19% after stent implantation (p = 0.007) after a median follow-up period of 6 months (interquartile range: 3 to 55 months). The total complication rate for PTA was 10% versus 13% for stenting (p = NS).
Conclusions This study demonstrates significantly better outcomes in terms of restenosis after stent implantation versus PTA only, with comparable complication rates for these 2 options of interventional treatment of radiofrequency-induced PVS. In summary, despite the lack of randomized studies, the present data and currently available published studies seem to favor stent implantation as a first-line therapy in patients with radiofrequency-induced severe PVS.
Pulmonary vein stenosis (PVS) is a commonly known complication after radiofrequency ablation (RFA) for treatment of atrial fibrillation (AF) (1). Because of thermal injury at the pulmonary vein (PV)–antral junction or even inside the PV ostia, an intimal proliferative process is triggered, leading to irreversible damage of the venous tissue, to myocardial collagen replacement, and subsequently to endovascular contraction (2). Replacing focal PV ablation with a circumferential antral ablation technique as well as the use of 3-dimensional mapping systems has lowered the mean incidence of PVS from 6.3% (estimated from publications between 1999 and 2004) (3) to currently 1% (1). According to the consensus statement on catheter and surgical ablation of AF, routine screening for PVS after RFA for AF is not recommended (4,5). Because of the lack of randomized data, specific treatment guidelines are not yet available. Percutaneous transluminal balloon angioplasty (PTA), stent implantation, and surgical repair are mentioned, but more detailed recommendations are not provided (5). With regard to the restenosis rates and procedure-related complications reported from single-center investigations, treatment of PVS remains a challenging field (6–12).
Therefore, the aim of this study was to analyze our single-center results and outcomes for interventional treatment of RFA-related PVS.
This investigation is an observational study. From January 2004 to September 2017, all patients with the diagnosis of PVS after RFA or cryoablation of AF were evaluated. The cohort also included patients who underwent index ablation in other hospitals (Figure 1).
The study was approved by the local ethics committee and was performed in accordance with the declaration of Helsinki (approval number 128/17).
Definition of PVS
Minimal luminal diameter of stenosed PVs was measured on multiplanar reformatted 3-dimensional angiographic datasets (using contrast enhanced computed tomography or magnetic resonance imaging [MRI]) of pre- and post-ablation imaging, and PVS was expressed as a percentage of post- versus pre-ablation luminal diameter reduction. The degree of PVS is classified as severe with luminal narrowing more than 70%, moderate for 50% to 70% narrowing, and mild for <50% narrowing (4,5).
Screening for PVS
From 2004 to 2007, transesophageal echocardiography was performed as routine screening for RFA-acquired PVS. The time of screening transesophageal echocardiography was between 6 and 12 months after PV isolation (PVI) or at time of new symptoms suggestive of PVS. In cases of abnormal findings such as accelerated peak flow (more than 1 m/s), subsequent computed tomography or MRI was performed to confirm and quantify PVS. Also in case of insufficient transesophageal echocardiographic quality but typical symptoms suggestive for PVS, a subsequent imaging diagnostic was initiated.
Because of the small number of detected PVS and the enormous logistic effort required, routine screening was stopped. Since 2008, diagnostics have been initiated only in case of clinical symptoms suggestive of PVS. Depending on availability and patient characteristics, computed tomography, MRI, and/or PV angiography was used as the imaging modality of choice for each patient. During the period after routine screening was stopped, asymptomatic PVS were detected as incidental findings in the context of reablation procedures.
Indication for treatment of PVS
The treatment indication was symptomatic PVS with narrowing more than 70% in a single stenosis or 60% in multiple ipsilateral stenoses. A lung perfusion scan using MRI was not performed in every case. Therefore, the evidence of a perfusion deficit was not obligatory.
PTA and stent implantation procedures were performed under deep propofol sedation. One case of PTA is illustrated in Figure 2. For vascular access, 2 femoral venous sheaths (11 and 8 F) and a 4-F arterial sheath for invasive RR-interval monitoring were used. After transseptal puncture, a steerable sheath (Agilis Sheath, St. Jude Medical/Abbott, St. Paul, Minnesota) was introduced into the left atrium. A 5-F multipurpose catheter was advanced over a 0.035-inch guidewire to the ostium of the affected PV. Contrast-enhanced venography was performed for confirmation and characterization of the stenosis. In cases of totally occluded PVs, angiography was undertaken by antegrade angiography via the pulmonary artery. In some cases, targeting of totally occluded PVs was supported by a 3-dimensional mapping system and a shell of the reconstructed left atrial anatomy (13). In these cases, existing computed tomographic or MRI data of the anatomy before the stenosis were helpful to plan the procedures of recanalization.
Intubation of PV and crossing of stenosis was supported by different guidewires depending on the degree of stenosis and the operator’s discretion (e.g., Supra Core [Abbott Vascular, Santa Clara, California], Glidewire [Terumo, Tokyo, Japan]). Balloon venoplasty was done by positioning an appropriately sized balloon at the point of maximal stenosis and inflating to a median pressure of 10 atm for 30 to 60 s (e.g., Admiral, Medtronic, Minneapolis, Minnesota). Success of dilation was assessed by contrast injection.
Usually, the first step of the treatment protocol was PTA. In case of insufficient dilation, a balloon with the next larger size or a different design (cutting balloon) was chosen. After challenging PTA with multiple balloons or early recoil of the PV, a stent was implanted even in the first procedure. Self-expanding stents were not applied, because of the high recoil forces of the stenotic PV. Because of the larger lumen of the PV, stents for coronary arteries were not used. Instead, stents for peripheral vascular treatment were used off-label. Residual narrowing of <10% to 20% was accepted as an endpoint of the procedure.
During the procedure, anticoagulation with a target active clotting time of 250 s was maintained by application of heparin. Intracardiac ultrasound guidance was not used.
The post-procedural anticoagulation regimen contained additional therapy with aspirin 100 mg/day for 4 weeks and clopidogrel 75 mg/day for 6 months in addition to current anticoagulation with vitamin K antagonists (target international normalized ratio 2.0 to 3.0) or direct oral anticoagulant agents in reduced doses.
The occurrence and degree of restenosis were registered by performing follow-up imaging. Restenosis of the previously intervened PV was defined as narrowing more than 70%.
All statistical tests were performed using SPSS version 20.0 (SPSS, Chicago, Illinois). Each categorical variable is expressed as number and percentage of patients. Continuous data are reported as mean ± SD. Follow-up time is given as median (interquartile range). The groups were compared using the chi-square test or Fisher exact test for categorical variables and the Wilcoxon rank sum test for continuous variables. Survival free from restenosis was estimated by cross-tabulation because of the small number of events. A 2-sided p value < 0.05 was considered to indicate statistical significance.
From January 2004 to September 2017, a total of 11,103 ablation procedures with PVI were performed at our institution, consisting of 10,971 catheter interventional RFAs and cryoballoon ablations, 115 intraoperative RFAs, and 17 hybrid PVI procedures as a combination of catheter interventional endocardial and surgical epicardial ablation.
Overall, 105 patients with a total of 172 PVS were registered. Eighty-seven of the index ablation procedures leading to PVS (83%) were performed at our institution, and 18 patients (17%) were referred with the diagnosis of PVS from an external hospital (Figure 1). There were no cases of PVS after cryoballoon ablation.
Overall rate of PVS
The total PVS rate for ablation of AF was 0.78% (87 of 11,103) and 0.74% (81 of 10,971) if surgical cases were excluded. There was no difference between the PVS rate during the transesophageal echocardiographic screening period in comparison with the period without routine screening (9 of 1,223 [0.74%] vs. 72 of 9,880 [0.73%], p = NS).
Treatment for PVS
Thirty-nine of 105 patients (37%) with PVS were treated with a total of 84 catheter interventional procedures. The distribution was 68 PTAs (81%) and 16 stent implantations(19%). Seven of the 16 stent implantations (44%) were done as primary stenting. The remaining 9 (56%) were stent implantations in a second instance after PTA with restenosis or in 1 case for acute management of PTA-induced PV rupture.
The baseline characteristics of the patients are shown in Table 1. The patients were predominantly men (60%) and had a mean age of 62.1 ± 9 years. The patients in the PTA group presented more frequently with arterial hypertension and had significantly higher CHA2DS2-VASc scores. The other cardiovascular risk factors and the left ventricular ejection fraction were well balanced between the groups.
Degree of severity and distribution of PVS
In total, the mean degree of stenosis was 92.6 ± 12.3% with no significant difference between the PTA and stent groups (92.2 ± 12.3% vs. 94.6 ± 12.7%, p = NS) (Table 2). The left inferior PV was most often affected, at 35%, followed by the right inferior PV at 26%, the left superior PV at 24%, and the right superior PV at 15%. There was no significance for the distribution of the affected PV between the groups.
Symptoms of patients with PVS
Thirty-five of the 39 treated patients (90%) reported at least 1 symptom related to PVS. The most frequent findings leading to the diagnosis of PVS were progressive dyspnea in 31 patients (79%), increase of pulmonary artery pressure in 13 (33%), and hemoptysis in 10 (26%). Mean time to diagnosis after index PVI was 10.2 ± 8.0 months.
PTA and stent procedures
The 68 PTA procedures were supported by a median balloon size of 10 × 20 mm with a median peak pressure of 10 atm (Table 3). In 9 cases, drug-eluting balloons were used, and in 2 cases, cutting balloons were used. The distribution was 54 (79%) as first PTA, 10 (15%) as second PTA, and 2 (3%) for each third and fourth PTA.
Regarding the 16 procedures with stent implantation, the median stent size was 7 × 20 mm. In 3 cases (19%), drug-eluting stents (DES) with a maximal diameter of 5 mm were implanted. Thirteen PVS (81%) were treated using bare-metal stents, whereas 8 were treated using polytetrafluoroethylene-covered stents.
The total complication rate for PTA was 10% versus 13% for stenting (p = NS) (Table 4). Major events in the PTA group were 2 wire-induced PV perforations with tamponade treated by pericardiocentesis and 2 balloon-induced PV ruptures with tamponade treated by emergency stenting and pericardiocentesis in 1 case and urgent surgical repair in the other case.
As for major events in the stent group, an acute stent thrombosis complicated by stroke occurred followed by intracerebral bleeding under systemic lysis therapy. Because of anatomic issues, 1 stent dislocated into the left atrium and was retrieved by snare. No patients died.
Imaging during follow-up was completed in 36 of the 39 patients (92%). Three patients from the PTA group did not consent to imaging, because of freedom from symptoms.
Restenosis occurred in 36 of 68 patients (53%) after PTA versus 3 of 16 patients (19%) after stent implantation (p = 0.007), with a median follow-up period of 6 months (interquartile range: 3 to 55 months) after the last procedure, without any difference between the groups. All patients with stent restenosis had stent diameters ≤8 mm. Regarding the PTA procedures, the restenosis rate was 29 of 54 (54%) after first PTA and 5 of 10 (50%) after second PTA. In the subgroup of 16 patients (13 PTA and 3 stenting), for the most part, no further interventions were performed, because of reductions of clinical symptoms despite persistent PVS. But we also observed severe cases that ended in surgical interventions such as lobectomy.
After follow-up, all patients without restenosis were asymptomatic due to symptoms suggestive for PVS.
Catheter intervention represents an established treatment strategy for RFA-acquired PVS. Our analysis demonstrates significantly higher restenosis-free survival after stent implantation versus PTA but comparable complication rates for percutaneous treatment of radiofrequency-induced PVS.
To date at our institution, the overall rate of RFA-acquired PVS after approximately 11,100 AF procedures is 0.78%. This PVS rate reflects a comparable level to the reported data from a worldwide survey as well as single-center experience (1,14).
The left-sided PVs are most commonly affected, as already known from single-center data regarding intervention for PVS (6–8,11). This may be a consequence of anatomic circumstance, with the small diameter of the left inferior PV and the cranial position of the left superior PV with the steep coumadin ridge leading to ablation sites closer to the ostium (15).
Interventional treatment for PVS
The overall restenosis rate was 53% for PTA versus 19% for stent implantation after a median follow-up period of 6 months (interquartile range: 3 to 55 months), whereas the complication rates did not differ significantly, at 6% for major events. Degree of stenosis and anatomic distribution of the affected veins did not differ between the groups.
The largest published registry on this topic reported on 124 patients with a total of 178 interventional treated veins (11). The 3-year restenosis rate was 49% of patients after PTA and 25% after stent implantation. The study of Prieto et al. (8) noted restenosis rates of 72% after balloon venoplasty and 33% after stent implantation in a cohort of 44 patients.
A meta-analysis including 2 additional studies with a total of 315 patients showed that balloon venoplasty was associated with a higher risk for restenosis in comparison with stenting (risk ratio: 2.18; 95% confidence interval: 1.64 to 2.89; p < 0.001), whereas the procedure-related complication rate was equal (16). Furthermore, the duration between index PVI for AF and interventional treatment for PVS did not affect the success of therapy (16).
The pathophysiological effect of restenosis after PTA is an elastic recoil and happens typically at an early stage after angioplasty. In comparison, restenosis after stent implantation occurs at a later stage because of neointimal hyperplasia (11). Our data suggest no significant relationship between stent diameter and restenosis, whereby all patients with stent restenosis had stent diameters ≤8 mm. Neumann et al. (17) reported a stent size of ≤10 mm for restenosis.
With regard to the pathophysiological correlate for in-stent restenosis, DES might be of benefit. A retrospective single-center comparison of different stent types reported a 3-fold higher restenosis rate after DES versus bare-metal stent implantation. Despite similar PV diameter, the stent diameter differed significantly, likely because of undersized DES (12). Currently available DES have a maximal diameter of 5 mm. Previous studies have demonstrated the feasibility of stent overexpansion with a 6-mm balloon but showed mechanical alterations with cell enlargement, strut distortion, and an accordingly reduced drug delivery per unit wall (18).
Treatment for restenosis
Restenosis after PTA can be treated by repetitive angioplasty procedures without further narrowing of the PV diameter. Furthermore, in case of restenosis despite multiple PTA procedures, surgical treatment by PV patch augmentation remains an option. Our single-center experience with surgical repair by patch augmentation for complex cases of multiple RFA-induced PVS revealed a cumulative restenosis rate of 38% after a long-term follow-up of 60 ± 69 months (19).
For in-stent restenosis after stent implantation, PTA is the treatment option of choice. To date, there is only 1 case report of successful augmentation for in-stent restenosis, but long-term outcome remains unclear (20).
Finally, one important fact remains to be highlighted with regard to interventional treatment of PVS by angioplasty or stenting: the angioplasty balloons and stents used were not specifically developed for treatment of PVS. The available tools for coronary artery interventions are too small for the PV diameter. Therefore, balloons and stents for peripheral vascular interventions with larger diameters are usually used off-label (6–13). Nevertheless, the lengths of these stents carry the risk for compromising branch vessels or overlapping into the left atrium.
Furthermore, there are currently no randomized data or clear guideline recommendations for treatment of PVS available. The Heart Rhythm Society consensus statement from 2012 suggests using PTA for severe stenosis and stent implantation in case of inadequate acute results or restenosis, and the role of surgery is discussed only for failed interventional treatment (4). There is no recommendation for patients with multiple PVS or how to deal with restenosis after interventional treatment. The latter consensus statement from 2017 mentions PTA, stent implantation, and surgical repair but does not provide more detailed recommendations (5).
This study was performed at a single high-volume center. As such, these data may not be easily applicable to other centers. The study design was retrospective, without randomization for the different treatment strategies. Because of the nonrandomized data, bias might be present in patient selection; furthermore, stent implantations were performed predominantly as second-line therapy after failed angioplasty, which may have contributed to the more favorable outcomes in the stent group. Regarding the PVS rate, there might have been an underestimation due to the exclusively symptom-based approach for diagnostic efforts since 2008. Because of the small number of DES used, a comparison between bare-metal stents and DES is not appropriate. The data for rates of restenosis may remain imprecise because of the variable timing of follow-up imaging. A Kaplan-Meier analysis for survival free from restenosis was not applicable, as there were only 3 restenosis events in the stent group.
Despite the lack of randomized studies, our data and currently available published research seem to indicate that stent implantation should be the first-line therapy in patients with RFA-induced severe PVS. To minimize restenosis rate, stents with diameters ≥8 mm should be preferred. In our experience, surgical repair can be considered in selected patients with 3 or more severe PVS or complex cases with prior failed PTA. There is a strong need for a prospectively randomized comparison of these 2 interventional treatment options for PVS following AF ablation procedures and for adequate products meeting the needs for PV stenting with short-length, large-diameter stents.
WHAT IS KNOWN? Data from a single-center retrospective analysis report the superiority of stenting in comparison with PTA for treatment of ablation-induced PVS.
WHAT IS NEW? Analysis of our single-center experience confirms this observation and provides further evidence in support of stenting as the first-line therapy for treatment of PVS.
WHAT IS NEXT? Randomized controlled trial data are needed to draft a clear recommendation for treatment of ablation-induced PVS.
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- atrial fibrillation
- drug-eluting stent(s)
- magnetic resonance imaging
- percutaneous transluminal balloon angioplasty
- pulmonary vein
- pulmonary vein isolation
- pulmonary vein stenosis(es)
- radiofrequency ablation
- Received February 5, 2018.
- Revision received April 2, 2018.
- Accepted May 8, 2018.
- 2018 American College of Cardiology Foundation
- Cappato R.,
- Calkins H.,
- Chen S.A.,
- et al.
- Taylor G.W.,
- Kay G.N.,
- Zheng X.,
- Bishop S.,
- Ideker R.E.
- Rostamian A.,
- Narayan S.M.,
- Thomson L.,
- Fishbein M.,
- Siegel R.J.
- Calkins H.,
- Kuck K.H.,
- Cappato R.,
- et al.
- Calkins H.,
- Hindricks G.,
- Cappato R.,
- et al.
- Packer D.L.,
- Keelan P.,
- Munger T.M.,
- et al.
- Di Biase L.,
- Fahmy T.S.,
- Wazni O.M.,
- et al.
- Holmes D.R. Jr..,
- Monahan K.H.,
- Packer D.
- Fender E.A.,
- Widmer R.J.,
- Hodge D.O.,
- et al.
- ↵Fink T, Schlüter M, Heeger CH, et al. Pulmonary vein stenosis or occlusion after catheter ablation of atrial fibrillation: long-term comparison of drug-eluting versus large bare metal stents. Europace 2018. In press.
- Hilbert S.,
- Sommer P.,
- Bollmann A.
- Teunissen C.,
- Clappers N.,
- Hassink R.J.,
- van der Heijden J.F.,
- Wittkampf F.H.,
- Loh P.
- Ho S.Y.,
- Cabrera J.A.,
- Sanchez-Quintana D.
- Buiatti A.,
- von Olshausen G.,
- Martens E.,
- et al.
- Ng J.,
- Foin N.,
- Ang H.Y.,
- et al.
- ↵Schoene K, Sommer P, Arya A, et al. Complex cases of acquired pulmonary vein stenosis after radiofrequency ablation—is surgical repair an option? Europace 2018. In press.
- Van Putte B.P.,
- Smith T.,
- De Kroon T.L.,
- Damiano R.J.