Cool enough? Lessons learned from cryoballoon‐guided catheter ablation for atrial fibrillation in young adults

Cryoballoon (CB)‐guided ablation of atrial fibrillation (AF) is established in symptomatic AF patients. This study sought to determine the safety and efficacy of CB pulmonary vein isolation (PVI) in young adults.


| INTRODUCTION
Pulmonary vein isolation (PVI) is considered as a first-line treatment in patients suffering from paroxysmal atrial fibrillation (AF) and is regularly performed in patients with drug-refractory AF. 1 Besides the conventional approach using radiofrequency (RF) energy, cryoballoon (CB) has emerged as a promising technique for successful and durable PVI. 2 In this context, the second generation CB (Arctic Front Advance™; Medtronic Inc., Minneapolis, MN, USA) was designed to further improve the acute efficacy of this single-shot device for PVI. While the general AF population tends to be old, a substantial part of AF patients reports symptoms suggesting AF or documented tachycardia episodes at earlier ages. 3 Furthermore, the pathogenesis of AF might differ between young and older AF patients. In the latter, the underlying pathophysiology can be related to comorbidities, such as left ventricular dysfunction or coronary artery disease, resulting in left atrial (LA) fibrosis and remodeling. 4 In young patients, diverse mechanisms including trigger foci in or in close relationship to the pulmonary veins (PVs), foci in the right and left atrium, as well as additional underlying arrhythmias may be responsible for the presence and maintenance of AF. [5][6][7] The long-term effect of PVI in young adults on arrhythmia progression remains unclear. However, ablation strategies in younger AF patients in general consist of performing a PVI after ruling out supraventricular tachycardia (SVT)-triggered AF. Despite its limitation regarding ablation strategies beyond PVI, a single-shot device is a very useful tool to reliably create continuous lesions around the PVs. Based on these assumptions, personalized paths in AF management, especially in young patients, are warranted and deserve further investigation.
Therefore, this study sought to determine the acute and long-term effect of CB-guided PVI in young adults and to identify predictors of AF/atrial tachycardia (AT) recurrence following catheter ablation in this specific patient cohort.

| METHODS
This observational single-center analysis consists of symptomatic AF patients aged <45 years and scheduled for CB-guided PVI at our institution between 2012 and 2018. Written informed consent was obtained from each patient before the procedure. The study was reviewed and approved by our local institutional review board and conducted in accordance with the Declaration of Helsinki. In all patients, documented AF by 12-lead ECG, Holter ECG or during continuous monitoring via an implanted device was a prerequisite for catheter ablation. All patients were routinely scheduled in our outpatient clinic 3, 6, and 12 months following the intervention.
Thereafter, structured telephone interviews were scheduled. If any history of AT/SVT was suspected, an electrophysiologic (EP) study was conducted before AF ablation and AT or SVT ablation was performed, if necessary. The CB ablation was then indicated in case of a persistence of AF following a successful SVT-ablation or in case of failure to induce another tachycardia during the EP study.
Transesophageal echocardiography was routinely performed before ablation to rule out LA thrombus formation. Furthermore, ECG-guided contrast enhanced computer tomography or MR imaging of the LA was utilized to identify variants in PV or LA anatomy and to determine the ostial PV diameter. If already present, oral anticoagulation with phenprocoumon was continued with a target international normalized ratio of 2.0-2.5. Direct oral anticoagulants (DOACs) were discontinued on the day of the procedure and resumed the same day after ruling out pericardial effusion. Oral anticoagulation was continued or stopped 3 months after the ablation procedure according to the patients' CHA 2 DS 2 -Vasc score. Antiarrhythmic drugs following the intervention were prescribed, if indicated.
As reported previously, [8][9][10] all procedures were performed under conscious sedation using propofol, fentanyl, and/or midazolam. Intravenous heparin was administered to maintain an activated clotting time of 300 s throughout the whole procedure. A multipolar mapping catheter (Achieve™; Medtronic Inc.) was introduced via the steerable sheath into each PV for verification of entrance and exit block. During freezing of the right-sided PVs, the phrenic nerve function was monitored by continuous pacing maneuvers. All patients were followed in our outpatient's clinic 3, 6, and 12 months after the index procedure. All cardiac devices were interrogated and screened for AF episodes if such a device was implanted. At each visit, they were asked about arrhythmia-related symptoms or discomfort during respiration. Moreover, a 72-h Holter ECG was regularly performed in all patients. Following a 3-month blanking period, recurrence was defined as any symptomatic episode of AT/AF lasting >30 s.

| Statistical analysis
Continuous data are presented as a mean ± SE. Categorical variables are reported as frequencies (percentages). We performed a Cox proportional hazards regression with gender, age, body mass index (BMI), type of AF, LA diameter and CHA 2 DS 2 -Vasc score included as covariates. Kaplan-Meier curves were used to estimate event-free survival. In addition, as part of our exploratory analysis, we conducted a variable selection technique (Elastic Net) to identify relevant risk factors. Moreover, univariable Cox models for every explanatory variable were fit. All calculations were performed with the statistical analysis software R (R version 3.5.1). The factors of the CHA 2 DS 2 -Vasc score were not included separately into the analysis model, except for the confounding variables of gender and age. Note that only a small number of recurrences occurred during the study and a relatively high number of factors was examined. Due to the relatively small sample size and the exploratory character of this study, we decided to counteract the problem of conducting multiple tests by using the Benjamini-Hochberg procedure, which controls the false discovery rate, not the family-wise error rate. Inspection of the Schoenfeld residual plots did not indicate a violation of the proportional hazard assumption.
For further exploration of the effect of age on event-free survival, we compared the results with a matched control group consisting of AF-patients older than 45 years.

| Patients' characteristics
A total of 93 consecutive patients were included. The mean age was 35 ± 7 years, 78% of them had paroxysmal AF, while the remaining had persistent AF. Congestive heart failure was present in 10% of the participants, and 9% had structural heart disease. The mean LA diameter in echocardiography was 36 ± 7 mm. Routine use of vitamin K antagonists or DOACs was recommended in 20 patients (19%). A total of 61% of patients were overweight (BMI ≥ 26) and 17% were obese (BMI ≥ 30). All patients' characteristics are listed in Table 1.

| Findings from previous EP study
An EP study was regularly performed before AF ablation. Atrioventricular (AV)-nodal reentrant tachycardia, a concealed AV reentrant tachycardia, and an ectopic AT were found in three, one, and three patients, respectively. Two of the ATs originated from the right atrium (crista terminalis and posterior CS ostium). Ablation of these arrhythmias was performed using a three-dimensional (3D) mapping system. The remainder was a left-sided tachycardia, which was initially treated with antiarrhythmic drugs. EP study failed to induce sustained tachycardia other than AF in remaining patients. Participants in our study had AF recurrences despite a successful treatment of their AT/SVT and were therefore scheduled to undergo an AF ablation procedure.

| Procedural data
The mean procedure duration was 111 ± 38 min, the mean fluoroscopy time was 10 ± 7, with a mean fluoroscopy dose of 1190 ± 1470 (cGy) × cm². The mean diameters of the PVs were measured as follows: left superior PV 17 ± 3 mm, left inferior PV 16 ± 4 mm, right superior PV 18 ± 4 mm, right inferior PV 18 ± 4 mm and in case of a left common trunk, which was present in 17 patients, we measured a mean diameter of 25 ± 9 mm. Acute procedural success, defined as entrance and exit block of all PVs, was achieved in all patients. The mean freezing duration per vein was 234 ± 92 s. The complication rate requiring intervention was low (1.92%): we observed one arteriovenous fistula and one mediastinal hematoma caused by a bleeding from the right pulmonary veins with self-tamponade. Of note, permanent palsy of the phrenic nerve and severe PV stenosis did not occur.

| Follow-up and freedom from arrhythmia recurrence
After 12 months of follow-up the recurrence rate was 8% (6/93).
Regarding the whole follow-up period of 2.6 ± 2 years (median: The remainder maintained the same type of AF after recurrence compared with their baseline.

| Predictors for arrhythmia recurrence after successful PVI
In addition to the Cox models, we also performed an exploratory analysis of our data using a variable selection technique (Elastic Net) to further identify risk factors for recurrence. Using this penalized variable selection method, male sex, and persistent AF were identified as risk factors for recurrence in addition to an enlarged LA diameter.

| Comparison with historical control group
We performed a propensity matched analysis using logistic regression of age on relevant clinical parameters with a historical control group consisting of CB patients older than 45 years. Characteristics of the control group are depicted in Table 4. The difference of the LA diameter was accepted given the fact that both values are within the normal range.
The recurrence rate in the control group was 27% in contrast to a recurrence rate of 15% in the younger patients. Increasing age was BERGAU ET AL. The recurrence-free survival within the first 12 months following the index procedure is depicted in Figure 2. Due to a high number of censored results, the logrank test did not reach statistical difference (p = . 16), however, the graph shows the statistical trend toward a beneficial outcome in the young.

| Major findings
In this observational study we analyzed the efficacy and safety of PVI using the second-generation CB in young adults suffering from AF.
We report three major findings: First, PVI using the CB technique is safe and effective in young patients. Second, short-and long-term success rate by means of freedom from any AF/AT recurrence following CB-guided PVI is relatively high (∼80%) and may therefore be suggested as a first-line ablation approach in this specific patient cohort. Third, young AF adults with findings suggestive of LA remodeling, such as a dilated LA diameter, are more likely to develop recurrent AF/AT following CB-guided PVI.

| Trigger arrhythmias, mechanisms, and substrates in young AF patients
An EP study was performed before AF ablation in most patients.
An SVT was detected and successfully ablated in 7% of patients. Supraventricular arrhythmias triggering the initiation or maintenance of AF is a relatively common phenomenon in young adults and has been reported in up to 39% of young AF patients. 11 Although it is still possible, that non-PV triggers were responsible for AF initiation in our patient cohort, a PV-dependent mechanism is the most likely mechanism. This might be a possible explanation for the acute and long-term success rate of CB ablation in these patients.

| Procedural considerations performing CB-guided PVI in young adults
Focusing on procedural data, procedure duration and freezing cycle length was comparable to other clinical studies, whereas the fluoroscopy time was lower than usually reported. This might be due to the widespread use of ICE to guide catheter ablation in our T A B L E 1 Baseline characteristics and comparison of patients with and without recurrence and long-term outcome by means of freedom from AF/AT recurrence, the isolation of the PVs using a CB with a standardized size appears to be safe and feasible in young patients irrespective of the individual PV anatomy. This might be an advantage in contrast to RF-guided ablation, in which the PV anatomy can significantly influence procedural success. 16 The overall complication rate was low, with no phrenic nerve palsy observed in our cohort. The frequency of this complication has been reported ranging from 4% to 13%. 13,17 The reason for this low rate in our study is not entirely clear. One explanation could be that the phrenic nerve is more resistant to the effects of freezing in younger patients or that it recovers quickly after a transient or asymptomatic palsy immediately after abortion of the freeze cycle.
Furthermore, no PV stenosis was observed in our study.
The mechanism of PV stenosis following CB ablation is caused by a combination of cellular damage, vascular damage, and immunologic phenomenon. 18 According to the literature, the incidence of PV stenosis following CB ablation ranges between 0% and 7%. 13,19 PV stenosis were related to a larger ostial diameter and a more distal application site in a study by Narui et al. 20 There might be two possible explanations for this observation in our study. First, the utilization of ICE ensures a proximal application site in most freezing cycles. Second, considering that most PV stenosis following CB ablation remain asymptomatic, 21  the presence of LA low-voltage areas significantly affected the success of CB-guided PVI.
A potential limitation of our analysis is the absence of evaluation of LA fibrosis extent in our patients before catheter ablation.
Our analysis demonstrates that in young AF patients with a low arrhythmia burden and few comorbidities, a high success rate of AF ablation can be expected using a single-shot device, even without additional information from electroanatomical mapping. Irrespective of this observation, it remains of utmost importance to identify other trigger arrhythmias causing or maintaining AF in young patients. Our results fill the gap of evidence in terms of efficacy, safety, and long-term success of CB-guided AF ablation, exclusively in young patients.

| CONCLUSION
PVI using the second generation CB is safe and effective in young adults. A general utilization of a single-shot device as first-line therapy in this certain AF population might be considered, warranting further prospective evaluation.