Author + information
- Received December 6, 2018
- Revision received March 26, 2019
- Accepted March 26, 2019
- Published online May 20, 2019.
- Shaojie Chen, MD, PhD∗ (, )
- Boris Schmidt, MD∗ (, )
- Stefano Bordignon, MD,
- Laura Perrotta, MD,
- Fabrizio Bologna, MD and
- K.R. Julian Chun, MD∗ ()
- Cardioangiologisches Centrum Bethanien, Frankfurt Academy for Arrhythmias, Medizinische Klinik III, Agaplesion Markus Krankenhaus, Frankfurt am Main, Germany
- ↵∗Address for correspondence:
Dr. Shaojie Chen, Dr. Boris Schmidt, OR Dr. K. R. Julian Chun, Cardioangiologisches Centrum Bethanien, Medizinische Klinik III, Agaplesion Markus Krankenhaus, Wilhelm-Epstein Strasse 4, 60431 Frankfurt am Main, Germany.
Objectives This study sought to evaluate the durability of pulmonary vein isolation (PVI) after 2 different freeze durations by using time-to-effect guided (ICE-T) second generation cryoballoon (CB2) ablation strategy in patients with atrial fibrillation (AF) undergoing repeat procedure.
Background CB2 represents a powerful technology for PVI. Recently, the ICE-T CB2 ablation strategy targeting a 240-s single freeze demonstrated fast and efficient PVI. To further optimize safety and efficacy, a shortened 3-min freeze duration has been suggested, but PVI durability remains unclear.
Methods Between May 1, 2013 and December 31, 2017, all CB2 ablations followed the ICE-T concept (target freeze: 240 s or 180 s). Patients undergoing a second procedure for arrhythmia recurrence were analyzed. Two groups were defined based on the index freeze duration (group A: 240 s vs. group B: 180 s). In all repeat procedures a 3-dimensional left-atrial map was obtained. Durability of PVI and localization of conduction gaps were compared.
Results Of 788 total patients, 106 (13%) underwent a second procedure (group A: 80 of 604 vs. group B: 26 of 184) after a mean of 377 days. There was no difference regarding PV occlusion and time-to-isolation in the index procedure between the 2 groups. No major complications occurred. During the second procedure, significantly more patients demonstrated durable isolation of all PV in group A (61% vs. 35%; p = 0.02) along with a significantly increased rate of PVI durability (88% vs. 69%, per vein; p < 0.001). Left-sided PV did significantly benefit from 240-s freeze (reconnection left superior PV: 6% vs. 27%; p = 0.004, left inferior PV: 14% vs. 39%; p = 0.006).
Conclusions The ICE-T ablation strategy is associated with a high rate of durable PVI in patients with arrhythmia recurrence. Target freeze duration of 240 s versus 180 s is associated with significantly increased lesion durability, particularly at left-sided PV, without increasing complications.
Pulmonary vein isolation (PVI) represents the cornerstone in catheter ablation of atrial fibrillation (AF) and can be considered as first-line treatment for patients with symptomatic AF (1). Cryoballoon (CB) PVI, particularly after the introduction of the second-generation cryoballoon (CB2), has become an established ablation approach providing similar clinical efficacy compared with the gold standard radiofrequency ablation (2). Importantly, the CB2 represents a powerful ablation tool that was associated with improved outcome but increased risk for collateral damage (3–6). The increased cooling power of CB2 was also confirmed by invasive electrophysiological PV remapping data (7,8). Initially, a freeze duration of 240 s followed by an empiric bonus freeze was recommended (9,10). However, different energy dosing protocols suggesting the following: 1) shorter freezes (180 s); 2) empiric omission of the bonus freeze; or 3) titrating energy based on an individualized time to isolation (TTI)-guided PVI approach have been introduced (11–14). The first randomized CB energy dosing trial (time-to-effect–guided freeze [ICE-T] trial) followed the TTI-guided PVI strategy with a target freeze duration of 240 s, omitting empiric bonus freezes depending on the observed TTI (≤75 s). This strategy resulted in faster AF ablation without compromising the favorable clinical outcome (15). However, PVI durability of ICE-T–guided CB2 lesions remains unknown. Moreover, no remapping data of ICE-T–guided PVI targeting 180-s versus 240-s freeze duration have been reported. Therefore, we aimed to investigate PVI durability of ICE-T–guided CB2 ablation (target freeze duration: 180 s vs. 240 s) in patients undergoing a repeat procedure due to atrial tachyarrhythmia (ATa) recurrence.
All AF ablation procedural data has been prospectively collected in the Cardioangiologisches Centrum Bethanien database since May 2010. In May 2013, the individualized TTI-guided PVI concept was introduced to clinical routine. In all patients, the single big CB2 (28-mm) technique was utilized. In consecutive patients, target ablation time was initially set to 240 s followed by 180 s. Patients undergoing a repeat electrophysiological procedure due to recurrent ATa were identified from the database and grouped according to their index ablation strategy (group A: 240 s vs. group B: 180 s). All patients provided written informed consent and the study was reviewed by the institutional review board.
Index CB2 ablation
All procedures were performed by the same experienced operators. In all patients, the ICE-T strategy following the individualized TTI-guided dosing concept was used, and in a subset of patients, the target freeze duration was reduced from 240 s to 180 s. The principle of the simplified single big CB technique has been described elsewhere (16,17). In brief, after a single trans-septal puncture (BRK-1 needle, 8.5-F SL1 trans-septal sheath, St. Jude Medical, Little Canada, Minnesota), PV angiographies were performed. The CB delivery sheath (12-F, Flex Cath Advance, Medtronic, Minneapolis, Minnesota) was advanced into the left atrium (LA). The 28-mm CB2 was used in conjunction with a multipolar spiral catheter (Achieve, Medtronic) (Arctic Front Advance—CB2, Medtronic), allowing registration of PV conduction and determination of the TTI.
PV occlusion was assessed by contrast medium injection through the distal CB lumen. A bonus freeze after PVI was only delivered if TTI was >75 s. The freeze was terminated if the TTI was longer than 90 s, and a better CB position and PV occlusion were perused. During ablation of both septal PV, an octapolar diagnostic catheter (7-F ParaHis, Biosense Webster, Irwindale, California) stimulated the phrenic nerve (PN) from the superior vena cava (12 V, 2.9 ms, 1,200 ms) to identify PN dysfunction early. The diaphragmatic compound motor action potential was monitored. In case of compound motor action potential amplitude reduction of more than 30% or weakness or loss of PN capture, the freeze was prematurely terminated. Transient PN palsy was classified if PN dysfunction resumed before hospital discharge at day 2 after the procedure. If PN injury remained at hospital discharge, persistent PN palsy was diagnosed. A temperature probe (SensiTherm, St. Jude Medical; or Circa Probe, CIRCA Scientific, Park City, Utah) was inserted into the esophagus to monitor the luminal esophageal temperature. The luminal esophageal temperature cutoff temperature was set to 15°C (18). All relevant procedural data have been prospectively collected.
Repeat ablation procedure
Patients were offered a repeated electrophysiological procedure if they experienced symptomatic and electrocardiographic documented ATa (>30 s) recurrence. Two groups were defined based on the index CB2 target freeze duration (group A’: 240 s vs. group B’: 180 s). All repeat procedures were performed under conscious sedation (midazolam, propofol, and fentanyl boluses followed by a propofol infusion). After double trans-septal puncture, a 3-dimensional electroanatomical LA map (CARTO 3, Biosense Webster) was performed using an irrigated-tip radiofrequency catheter (ThermoCool SF, ST SF NaviStar, Biosense Webster). Selective PV angiograms were performed and both ipsilateral PV ostia were tagged in the LA CARTO map as previously described (2,7). If patients presented in sinus rhythm a decapolar spiral catheter (LASSO, Biosense Webster) was positioned within the proximal PV ostium to assess potential LA–PV reconduction. If LA–PV reconduction was present, the PV spike sequence was analyzed to identify the earliest PV activation site. Irrigated tip radiofrequency ablation was guided by local map potential along the tagged PV ostium and PV activation sequence. Gap location was defined as the site of successful reisolation of the PV or the site of clear change in the PV electric activation pattern as described before (19). Ipsilateral PV were divided into 4 quadrants—superior, inferior, anterior, posterior—to categorize the gap location. In patients presenting with atrial tachycardia (AT), the underlying AT mechanism was elucidated combining 3-dimensional electroanatomic mapping with conventional electrophysiological information such as entrainment maneuvers. After AT termination, PV gap assessment and re-PVI was performed in sinus rhythm. In patients presenting in AF, re-PVI was performed after cardioversion. In selected patients, individual substrate ablation was performed based on LA bipolar voltages assessed in sinus rhythm.
After the repeat ablation procedure, all patients were scheduled in our outpatient clinic at 3, 6, 9, and 12 months and in 6-month intervals thereafter and obtained a 72-h Holter electrocardiograph. ATa recurrence was defined as documented episodes lasting >30 s.
The primary endpoints were PV reconduction and localization of PV conduction gaps after index CB2 ICE-T–guided PVI comparing both groups (240-s vs. 180-s target freezing time).
The secondary endpoints were index and second ablation procedural parameters, complications, and time of second AF ablation procedure.
Continuous variables were described as mean ± SD. An independent Student's t-test or Mann-Whitney U test was used for continuous variables as appropriate. Categorical variables were described as numbers or proportions and analyzed using the chi-square test or the Fisher exact test when appropriate. The PVI durability was evaluated per individual PV—left superior PV, left inferior PV, right inferior PV, and right superior PV—irrespective of common root anatomy. A p value of <0.05 was considered statistically significant. The statistical analyses were performed using SPSS version 17.0, (SPSS Inc., Chicago, Illinois).
Between May 2013 and December 2017, a total of 788 consecutive patients underwent CB2 AF ablation using the ICE-T concept (target freeze of 240 s: n = 604 patients; target freeze of 180 s: n = 184 patients). A total of 106 patients (13%) underwent a repeat ablation procedure after a mean follow-up of 377 (range 140 to 512) days (group A: n = 80 vs. group B: n = 26). Table 1 summarizes patients baseline characteristics (n = 106) who underwent second AF ablation including PV remapping. No difference in the patient baseline characteristics was noted between groups.
Durability of PVI
During the second procedure, a total of 424 PV (group A: 320 PV) and (group B: 104 PV) were remapped using a spiral catheter to assess PVI durability. A significantly higher rate of PVI durability was observed in group A (280 of 320 PV: 87.5%) compared with group B (72 of 104 PV: 69.2%; p < 0.0001). Details regarding the PV reconduction pattern per PV and per patient are summarized in Table 2, and rates of PV reconnection were shown in the left panel of the Central Illustration.
PV Conduction Gap Pattern
Table 3 and the Central Illustration (right panel) summarize the PV Reconduction Gap Location Distribution identified in the second procedure. For the right PV, the most common gap site in ICE-T Group A was located in inferior (16.3%), followed by superior (12.5%), posterior (1.3%), and anterior (0%); the most common gap site in ICE-T Group B was located in superior (23.1%), followed by inferior (15.4%), anterior (7.7%), and posterior (7.7%).
For the left PV, the most common gap site in ICE-T group A was located in anterior (11.3%), followed by superior (3.8%), inferior (3.8%), and posterior (1.3%); the most common gap site in ICE-T group B was located in anterior (34.6%), followed by superior (15.4%), inferior (11.5%), and posterior (3.8%). Figure 1 shows an example of PV reisolation in the remapping procedure.
Patients and durable isolation of all PV
Table 4 shows durability of PVI according to target freeze duration (240 s vs. 180 s). After a mean follow-up of 377 (range 140 to 512) days, durable isolation of all PV was significantly more often present in Group A versus Group B (61.3% vs. 34.6% of patients; p = 0.02). This was driven by lower rates of 2-PV, 3-PV, and 4-PV reconnection in ICE-T Group A.
Univariate analysis for PV reconnection
Table 5 shows the univariate analysis between patients with or without PV reconnection. Based on clinical and procedural variables, the target duration was the only significant predictor for PV reconnection (p = 0.018).
Table 6 summarizes the index procedural data of all remapped patients. In Group A, longer procedure (72.6 ± 21.6 min vs. 58.6 ± 23.5 min; p = 0.006) and LA time was observed than in Group B (61.1 ± 19.8 min vs. 49.3 ± 19.7 min; p = 0.009). The nadir CB temperature was significantly lower in Group A than in Group B (−48.5 ± 5.5°C vs. −44.8 ± 6°C; p = 0.005). There were no significant differences with regard to TTI, PV occlusion, number of freezes, and fluoroscopy time.
There were no major adverse events such cardiac tamponade, stroke/transient ischemic attack, PV stenosis, pulmonary bleeding, persistent PN palsy, or death in both groups. Transient PN palsy (5% vs. 3.8%; p = 0.809) and premature freeze termination due to reaching the 15°C esophageal temperature cutoff (16.3% vs. 11.3%; p = 0.756) was more common but not significant in group A.
The cornerstone of any AF ablation represents immediate PVI, which is not necessarily equivalent to a transmural lesion and long-term durable PVI (20). To the best of our knowledge, this presented remapping study sheds initial light into the black box of lesion durability following TTI-guided CB2 ablation. The main finding of our study provides evidence that CB2 aiming for a single freeze following the ICE-T strategy results in a high rate (83%) of durable PVI in patients undergoing a second AF ablation procedure. Importantly, a target freeze duration of 240 s compared with 180 s was associated with a significantly higher rate of durable PVI (88% vs. 69%; p < 0.001) (Central Illustration). The difference in PV reconnection pattern suggests a critical role of the additional 60-s freeze duration to improve lesions and transform functional block into a transmural lesion. As expected, the 240-s freeze protocol resulted in slightly longer procedure times but did not affect the benign complication profile.
Current practice of CB energy dosing
CB PVI has been established in routine AF ablation worldwide. The FIRE AND ICE (FIRE AND ICE: Comparative Study of Two Ablation Procedures in Patients With Atrial Fibrillation) study targeted a CB freeze duration of 240 s, which was followed per protocol by 1 empiric bonus freeze after isolation (2). This dosing protocol has become standard of care in many centers. However, the ideal cryothermal energy dosing balancing risks and benefits has been ill defined (9). This gap of evidence has become even more relevant after CB2 introduction with significantly improved cooling power (10,21,22) along with the observation of increased collateral damage (3–6). Interestingly, animal data showed that a single freeze (3 min) without a bonus application creates a transmural lesion in dogs (23). Therefore, to shift CB2 ablation toward greater safety, dosing protocols reducing target freeze duration to 180 s with or without an empiric bonus freeze have been introduced but have been lacking systematic PV remapping data (11,12). A different dosing concept titrates freeze duration with or without empiric bonus freezes based on TTI (15,24,25). The first randomized trial investigating TTI-guided ablation (ICE-T trial) resulted in fewer empiric bonus freezes and shorter procedure times without compromising the clinical outcome. Procedure-related complications were reduced (15). Recently, the clinical value of TTI-guided ablation has been confirmed in a second randomized trial (13). In addition, a short TTI has been identified as an independent predictor of freedom from ATa recurrence (15,26), underlining its value to guide CB2 AF ablation.
Dosing CB energy: Efficacy and safety
In an initial human study investigated 21 patients after an index “standard” CB2 ablation protocol (240-s freeze + empiric bonus freeze). Notably, 1 patient experienced substantial esophageal ulceration and bleeding. All patients were restudied after 3 months regardless of ATa recurrence, demonstrating a high rate of durable PVI (91%, 68 of 75 PV) (8). This observation is in line with subsequent studies using a similar “PVI + 1” (240-s) empiric dosing protocol. In these patients with ATa recurrence, invasive PV remapping demonstrated high rates of durable PVI (7,27,28). As expected, limiting target freeze duration to 180 s did impact CB nadir temperatures (group B: −44.8°C vs. group A: −48.5°C) and shortened both ablation and thawing time. This resulted in 14-min procedure time savings while risking a significantly greater chance for PV reconnection over time. Importantly, the extension of a cryothermal lesion is based on local balloon to tissue contact and on freeze duration (29). Thus, the left-sided anterior PV segment (the ridge to left appendage) did particularly benefit from increased freezing duration in group A to transform functional block into a transmural lesion (30,31).
Left-sided PV anatomy is typically characterized by a shorter inter-PV distance and thicker myocardial tissue (32,33). Both factors may explain the observed different PV reconnection pattern. The shorter inter-PV distance allows for a constant warming blood flow of the corresponding PV on the balloon, and therefore a longer freeze duration may be required to create a transmural lesion. According to our data, even a full 240-s freeze may not be sufficient enough to create permanent PVI in all anatomic settings and patients. In contrast, for right-sided PV, a 240-s freeze was associated with significantly improved chronic right superior PV isolation but no significant difference was observed for the right inferior PV. Whether the freeze duration after a pull-down maneuver needs to be prolonged or early termination due to esophageal cooling may have affected this finding remains open. In group A (240-s freeze) more but not significantly premature freeze terminations occurred due to early signs of potential collateral damage (15°C esophageal temperature: 16.3% vs. 11.3%; p = 0.756, transient PN palsy: 5% vs. 3.8%; p = 0.809). Importantly, no pulmonary bleeding was observed in both groups. Therefore, to preserve the favorable low complication rate, early detection of potential collateral damage by recording esophageal temperatures and performing PN pacing is essential.
Initial PV remapping study showed improved CB2 versus first-generation CB lesion durability (240 s + empiric bonus freeze) in patients undergoing a second AF ablation (7,8). In the present larger study (n = 106 patients), we found that the overall durability of PVI increased (83% per veins) by using the single freeze ICE-T–guided CB2 strategy. Furthermore, in group A we found a significantly higher rate of chronic PVI (88% per veins) in a second procedure. These results are reassuring with regard to the concept of ICE-T–guided PVI proving efficient lesion formation while omitting the empiric bonus application. Interestingly, the univariate analysis found that freeze duration was the only significant predictor for subsequent PV reconnection. This appears to indicate that given a proper use of CB2, total freeze duration represents a key parameter in obtaining a durable and transmural lesion.
First, the presented study is a nonrandomized, retrospective analysis based on registry data. To fully compare the 240-s and 180-s freeze protocols, the whole study population (both patients with or without AT/AF recurrence) should be remapped in a repeated procedure after the same follow-up. This was very difficult if not impossible to achieve, as remapping all patients even without AT/AF recurrences was not feasible in this cohort study. To minimize the bias, efforts have been made: All patients were consecutively enrolled, and the patients were blinded to the freeze duration. All procedures were performed by the same experienced operators, and the follow-up was carried out by physicians who were blinded to the freeze protocol. Second, the primary study endpoint focused on the long-term PVI durability assessed in remapping procedures. Clinical rhythm outcome of different dosing strategies was not investigated in this analysis; however, it is being investigated further.
The individualized TTI-guided CB2 ablation strategy is associated with a high rate of durable PVI in patients with ATa recurrence. Target freeze duration of 240 s versus 180 s is associated with significantly increased lesion durability, particularly at left-sided PV, without increasing complications.
COMPETENCY IN MEDICAL KNOWLEDGE: The individualized TTI-guided CB2 ablation strategy shows a high rate of durable PVI in patients with ATa recurrence. The TTI-guided 240-s freeze duration is associated with significantly higher rate of durable isolation of PV without increasing complications as compared with the 180-s freeze duration.
TRANSLATIONAL OUTLOOK: Further prospective multicenter randomized studies are necessary to determine the optimal freeze protocol of cryoablation for AF in terms of enduring PVI as well as clinical outcomes.
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 tachycardia
- atrial tachyarrhythmia
- second-generation cryoballoon
- time-to-effect–guided freeze
- left atrium
- phrenic nerve
- pulmonary vein
- pulmonary vein isolation
- time to isolation
- Received December 6, 2018.
- Revision received March 26, 2019.
- Accepted March 26, 2019.
- Metzner A.,
- Reissmann B.,
- Rausch P.,
- et al.
- Straube F.,
- Dorwarth U.,
- Hartl S.,
- et al.
- Ferrero-de-Loma-Osorio Á.,
- García-Fernández A.,
- Castillo-Castillo J.,
- et al.
- Pott A.,
- Kraft C.,
- Stephan T.,
- Petscher K.,
- Rottbauer W.,
- Dahme T.
- Chun K.R.,
- Stich M.,
- Fürnkranz A.,
- et al.
- Chen S.,
- Schmidt B.,
- Bordignon S.,
- et al.
- Kuck K.H.,
- Hoffmann B.A.,
- Ernst S.,
- et al.,
- for the Gap-AF–AFNET 1 Investigators
- Takami M.,
- Lehmann H.I.,
- Misiri J.,
- et al.
- Aryana A.,
- Kenigsberg D.N.,
- Kowalski M.,
- et al.,
- for the Cryo-DOSING Investigators
- Su W.,
- Aryana A.,
- Passman R.,
- et al.
- Aryana A.,
- Singh S.M.,
- Mugnai G.,
- et al.
- Kawaguchi N.,
- Okishige K.,
- Yamauchi Y.,
- Kurabayashi M.,
- Hirao K.
- Su W.,
- Coulombe N.,
- Kirchhof N.,
- Grassl E.,
- Wittenberger D.
- Chen S.,
- Schmidt B.,
- Bordignon S.,
- et al.
- Bordignon S.,
- Chen S.,
- Perrotta L.,
- et al.