Data collection adhered to the principles outlined in the Declaration of Helsinki (2013) and Good Clinical Practices. The study was approved by local institutional review boards (IRBs) and ethics committees at each participating center, and patients provided written informed consent prior to participation in the registry (Samsung Medical Center IRB 2019-02-061, Seoul National University Bundang Hospital B-1903/562-303 and Hallym University Sacred Heart Hospital 2019-04-025). The data, analytic methods, and study materials will not be available to other researchers. Specifically, patient data privacy within the Cryo Global Registry does not allow nor consent to data sharing with outside parties.

The Cryo Global Registry (NCT02752737) is an ongoing, prospective, multicenter post-market registry to evaluate outcomes on AF ablation procedures performed with the Arctic Front Advance Family of cryoablation catheters (Medtronic, Inc., Minneapolis, MN, USA). The current analyses aimed to describe patient baseline characteristics, procedural characteristics, and outcomes of CBA in the Korean sub-population within the Cryo Global Registry. Data regarding the ablation procedures were collected from 299 patients enrolled in 3 hospitals in Korea.

The registry has been previously described in detail by Chun et al.9) In brief, a global steering committee of international physicians oversees data quality, analyses, and publication milestones. Procedures performed in this study were performed according to standard-of-care at the time of data collection.

Patients in this analysis were consecutively enrolled between April 2019 to May 2020 at 3 hospitals in Korea. All patients ≥18 years old with a planned CBA procedure were eligible for inclusion in the registry and were not excluded based on pre-existing characteristics nor medical conditions. All patients received PVI ablation for the first time, and patients were classified by AF disease status, including: PAF (AF that terminates spontaneously or with intervention within 7 days of onset), PsAF (continuous AF that is sustained beyond 7 days and ≤12 months), or long-standing PsAF (LsPsAF, continuous AF >12 months).3) LsPsAF patients were pooled with PsAF patients in this analysis of 3 centers for statistical considerations because of sample size considerations.

CBA-PVI was performed according to local standard-of-care in Korea and in accordance with typical procedural techniques.7)8)9) In brief, a dedicated 15-F OD steerable sheath (FlexCath Advance Steerable Sheath; Medtronic, Inc.) was used to introduce a 23- or 28-mm CBA catheter (Arctic Front Advance; Medtronic, Inc.) into the left atrium (LA). The CBA catheter was maneuvered in the LA either over a J-tip guidewire or a dedicated inner-lumen (octopolar or decapolar) circular mapping catheter (Achieve or Achieve Advance; Medtronic, Inc.). The CBA catheter was inflated and advanced towards the antral surface of the pulmonary vein (PV). Freezing was initiated upon antral occlusion of the targeted PV. The number of cryoapplications and freeze duration were operator determined, and PVI was confirmed by entrance and/or exit block after CA.

Phrenic nerve monitoring (by diagnostic catheter pacing at the subclavian vein and diaphragmatic movement monitoring by manual palpitation and/or adjunctive diaphragmatic movement detection methods) was recommended during all right-sided PVI. Freezes were stopped upon detection of an attenuated diaphragmatic response. Pre- and intra-procedural imaging, esophageal temperature monitoring, and post-ablation chemical testing were determined by the operator. Additional ablation tools used (e.g., focal cryoablation and radiofrequency CA), as well as lesions adjunctive to PVI, were left to the discretion of the operator and were documented. Periprocedural anti-coagulation and AAD initiation/continuation were determined by the operator, and patients were discharged according to the hospital’s standard-of-care practice.

Patients were followed via telephone and/or in-office visits according to their hospital’s standard-of-care method. Patients were protocol-required to have at least one status visit at 12 months after the index CBA. The primary objective of the registry was to report efficacy and safety after the index CBA procedure throughout 12 months. The Korean patients that are represented in this analysis are a subset population from the larger Cryo Global AF Registry. Procedural efficacy was assessed by time-to-first ≥30-second recurrence of AF and atrial arrhythmias (defined as AF/atrial flutter [AFL]/atrial tachycardia [AT]) between a 90-day blanking period and 12-month follow-up. The method of rhythm monitoring was not dictated by the study protocol and could be conducted by any of the following methods, including: 12-lead electrocardiogram (ECG), Holter monitor, trans-telephonic monitor, insertable cardiac monitor, pacemaker, and/or implantable cardioverter defibrillator. The safety endpoint was the serious device and/or procedure-related adverse event rate as classified by the physician. Non-serious procedure-related adverse events and non-procedure related serious adverse events were also reported. Serious adverse events were defined according to international standards (ISO 14 155:2001) and included all events that led to death, or to a serious deterioration in health that resulted in either (a) a life-threatening illness or injury, (b) a permanent impairment in body structure or function, (c) in-patient or prolonged hospitalization, or (d) medical intervention to prevent life-threatening illness or injury.

Ancillary objectives were examined in this analysis including baseline characteristics, procedural characteristics, quality of life (QoL), arrhythmia monitoring, post-ablation medication management, and hospitalization rates. QoL (measured by 3-level EuroQol 5-dimensional questionnaire [EQ-5D-3L]) and predefined arrhythmia symptoms were monitored at baseline and 12-month follow-up. The EQ-5D-3L questionnaire measures 5 dimensions of health, including: 1) mobility, 2) selfcare, 3) physical activities, 4) pain and discomfort, and 5) anxiety and depression. Each question has 3 levels of response indicating no problem, some problem, or extreme problem. The resulting index score is on a scale ranging from 0 (least healthy) to 1 (most healthy). Class I/III AAD utilization was assessed at baseline, after hospital discharge for post-ablation care, and at the 12-month visit. The method and frequency of rhythm monitoring was documented throughout 12 months of follow-up, as well as the all-cause and cardiovascular (CV)-related hospitalization rates, and the repeat ablation rate.

Baseline characteristics and clinical data were summarized using the appropriate summary statistics. Continuous variables were summarized as mean and standard deviation, and categorical variables were summarized as counts and percentages. The Kaplan-Meier method was used to estimate the 12-month freedom from AF recurrence, all atrial arrhythmia recurrence, repeat ablation, and rehospitalization. Standard error was calculated with Greenwood’s formula. Separate log-rank tests were performed to assess the rate of AF recurrence and all atrial arrhythmia recurrence between PAF and PsAF subjects. Changes in QoL was assessed with a one-sample t-test. Values of p<0.05 were considered statistically significant. Statistical analyses were performed with SAS software version 9.4 (SAS Institute, Cary, NC, USA).

A total of 300 patients with AF were enrolled in 3 Korean hospitals. One patient was withdrawn by the investigator and exited the study prior to the index CBA; while 299 patients were treated according to standard practice. Table 1 describes baseline characteristics of the study population. Patients were on average 60±11 years of age, 75.3% male, 50.5% PAF, and diagnosed with AF for a mean of 2.3±2.8 years. PAF patients were diagnosed with AF on average 2.3±2.8 years, and PsAF patients were diagnosed for 2.4±2.8 years. The mean LA diameter was 43±7 mm (range 24–69 mm), and the left ventricular ejection fraction was 59±8%. A total of 62 (20.7%) patients had a history of prior AFL with 34 (22.5%) PAF patients and 28 (18.9%) PsAF patients. Seven patients (2.3%) had a history of prior AT with 6 (4%) in the PAF and 1 (0.7%) in the PsAF cohort. In addition, 3 (2.0%) patients in the PAF cohort had a previous AFL ablation vs. no patients in the PsAF cohort. Patients failed a mean of 1.2±0.6 AADs before the index-PVI ablation, and 4 patients (1.3%) were treated with first-line CBA.

All patients underwent index CBA using the 28-mm Arctic Front Advance cryoballoon. In 1 patient, a 23-mm cryoballoon was used in addition to a 28-mm balloon. Procedural characteristics are presented in Table 2. Mean total procedure time, LA dwell time, and fluoroscopy time were 76±21 minutes, 56±23 minutes, and 27±23 minutes, respectively. Preprocedural mapping with computed tomography or magnetic resonance imaging was performed in 191 (63.9%) patients. During the procedure, intracardiac echocardiography was performed in 52.5% of cases, and a PV venogram was done in 84.3% of cases. Intra-procedural 3D electroanatomical mapping was not used, and esophageal temperature was monitored in 195 (65.2%) patients. Phrenic nerve function was monitored in all patients by pacing and palpating the phrenic nerve in conjunction with additional monitoring methods such as Compound Motor Action Potential monitoring during right-sided PV CBA.

A total of 1,209 veins were treated with a mean of 1.5±1.0 freeze applications per vein. The average freeze duration was 167±54 seconds, and a mean nadir balloon temperature of −49.9±6.8°C was reported. There were no left common PV ostia treated with cryoballoon in this population. Acute PVI success was achieved in 98.0% of patients using only the Arctic Front Advance cryoballoon. No focal RFA or focal cryo touch-up applications were performed to achieve PVI. Overall, 72 (24.1%) patients underwent adjunctive non-PVI ablation during the index procedure. Of those, 66 patients underwent cavotricuspid isthmus (CTI) ablation, 29 (19.2%) PAF patients and 37 (25.0%) PsAF patients. In a total of 8 cases, other non-PVI and non-CTI ablations were performed, including 4 (2.6%) PAF patients and 4 (2.7%) PsAF patients. The non-PVI, non-CTI targets were LA AF triggers (1.3%, CBA), superior vena cava vein triggers (0.7%, CBA), left atrial anterior wall (0.3%, RFA), and a paroxysmal supraventricular tachycardia (0.3%, RFA).

For the effectiveness analysis, PAF patients were compared to PsAF patients. Arrhythmia recurrence is shown in Figure 1. The Kaplan-Meier estimate for freedom from a ≥30-second recurrence of AF at 12 months was 83.9% (95% confidence interval [CI], 76.9–88.9%) in PAF and 61.6% (95% CI, 53.1–69.0%) in PsAF (Figure 1A). Freedom from ≥30 secones AF/AFL/AT at 12 months was lower, with 75.2% (95% CI, 67.4–81.3%) in the PAF patient cohort and 56.7% (95% CI, 48.2–64.4%) in the PsAF patient cohort (Figure 1B). AF-free survival and atrial arrhythmia-free survival were significantly lower for the PsAF patients when compared to PAF patients (p<0.01 for both). The type of arrhythmia recurrence was AF in most of the PsAF patients (83.0%). In PAF patients, 18 patients (48.6%) had recurrent AF, while 29.7% experienced AT and 13.5% AFL (Figure 1C).

AAD continuation or prescription for post-ablation care was left to the discretion of the investigator. Of the 299 patients, 290 had AAD status reported at the index procedure discharge and 12 months. Overall AAD usage decreased from 82.4% at index procedure discharge (PAF: 75.0%, PsAF: 90.1%) to 44.8% at 12-month follow-up (PAF:38.5%, PsAF 51.4%; Figure 2A). In total, 98 patients had an atrial arrhythmia recurrence, and of those 29 (29.6%) patients were off AAD at 12 months (Table 3). During follow-up, 18 (18.4%) patients received a repeat ablation (6 patients off ADD and 12 patients on AAD at 12 months). Of the 192 patients that did not have an atrial arrhythmia recurrence, 131 (68.2%) were off AAD at 12 months, and none of those patients received a repeat ablation.

The method of rhythm monitoring was not dictated by the protocol. During the 12-month follow-up period, patients were seen in the clinic an average of 4.7±1.1 visits, including a study required 12-month visit, and the remaining visits were completed via standard of care follow-up. All patients were monitored for atrial arrhythmia recurrence at least once (Figure 2B). All patients received a 12-lead ECG during follow-up, and 80.3% patients had 3 or more 12-lead ECGs performed during follow-up. Holter monitoring was used at least once during the follow-up period in 61.2% of patients. In 26 patients (8.7%) with continuous monitoring devices, 74 data reviews were performed, and all these patients received an additional 12-lead ECG recordings according to standard-of-care.

A total of 62 total adverse events were reported in 51 patients (Table 4). Two serious device- and/or procedure-related events were reported in 2 patients (0.7%) within one day after the index procedure. In one patient, a vascular pseudoaneurysm was reported, and one incision site hematoma was observed in a second subject. Additionally, 5 non-serious procedure-related events (1.7%) were reported in 5 patients. Three of these events were phrenic nerve palsies, but none were classified as serious by the investigators. All phrenic nerve palsies were resolved without further clinical sequelae. In one patient, the phrenic nerve palsy was resolved at hospital discharge and the 2 others resolved at days 16 and 175 post- procedure (with a mean time to resolution of 64.0±96.4 days). Lastly, the investigators reported 55 other serious non-procedure related events (in 45 patients, 15.1%). Of those 55 serious events, 43 were related to atrial arrhythmia recurrence. Importantly, there were no atrioesophageal fistula, pericardial tamponade, or PV stenosis reported.

QoL, measured by the EQ-5D index score, improved significantly in the Korean patient cohort from 0.82±0.24 at baseline to 0.90±0.20 at 12-month follow-up (p<0.001). At baseline, 78.6% of patients reported to have experienced at least one symptom related to AF, and a distribution of symptoms is presented in Figure 2C. Patients most often experienced palpitations (58.3%) followed by dyspnea (17.9%) and dizziness (14.8%) at baseline. Symptom burden improved significantly after CBA, with 94.1% of patients being free of symptoms at 12-month follow-up.

Healthcare utilization after the index ablation was demonstrated using the Kaplan-Meier method for reporting the number of repeat ablations, all-cause hospitalizations, and CV-related hospitalizations throughout 12 months (Figure 3). CBA was effective in preventing repeat ablations, with 93.8% of patients being free from repeat ablations at 12 months. In total, 18 patients had a repeat ablation within the first year, and there were no patients that underwent a second repeat ablation. The estimate for freedom from first all-cause hospitalization post-ablation was 88.7% in this Korean cohort. Most hospitalizations were due to supraventricular arrhythmias (29/33), and freedom from the first hospitalization post-ablation due to CV-related causes was 89.7%.