Etiologic factors and long-term outcomes of catheter ablation of atrial fibrillation (AF) in young patients have not been well characterized. This study aimed to analyze the etiologic factors and outcomes of pulmonary vein isolation (PVI) in patients with young-onset AF (young-AF, defined as having first documented episode <45 years). Consecutive patients with young-AF undergoing PVI (n = 197) in 2 academic centers were enrolled and followed for 36.1 ± 24.5 months. A control group of patients with AF onset ≥45 years (n = 554) was included. The most frequent risk factors in young-AF were intensive exercise (25%), moderate-to-heavy alcohol consumption (23%), and familial AF (22%). Compared with patients with AF onset ≥45-year, patients with young-AF were more often men (82% vs 66%, p <0.001), had more frequently paroxysmal AF (81% vs 60%, p <0.001), had less left atrial dilatation (40.9 ± 6.2 mm vs 44.2 ± 7.2 mm, p <0.001), and had lower 4-year recurrence rate after last PVI procedure (22% vs 45%, p <0.001). In young-AF, structural heart disease (SHD) was the only independent predictor of recurrence. Patients with young-AF selected to undergo cryoballoon (CB) ablation were younger (35.0 ± 7.7 vs 36.6 ± 6.7 years, p = 0.035) and had less persistent AF (6% vs 24%, p = 0.004) and coronary artery disease (2% vs 7%, p = 0.02) compared with radiofrequency ablation. After excluding patients with persistent AF and SHD, there was no difference in single procedural success between radiofrequency or CB PVI (27% vs 17%, p = 0.11). In conclusion, patients with young-AF have diverse etiologies and high single and multiprocedural PVI successes. SHD is the only independent predictor of recurrence. In patients with young-AF, there is a selection bias for CB ablation.
Despite age being one of the most important risk factors for atrial fibrillation (AF), the prevalence of young-onset AF (young-AF) is rising. Previously described important differences in this population compared with older patients with AF include a lower likelihood of important underlying co-morbidities, aversion of lifelong medication intake, and more symptomatic clinical course. The previously reported risk factors for AF in the young population were diverse as depicted in Supplementary Table 1 . In the absence of specific guidelines for young-AF, the management of these patients has been extrapolated from trials of older patients. So far, smaller studies have investigated the clinical outcome of catheter ablation in young patients with AF, as summarized in Table 1 . This study was to analyze the specific etiologic factors and long-term outcomes of pulmonary vein isolation (PVI) in a large population of patients with young-AF in 2 academic European centers.
Consecutive patients undergoing PVI between January 2012 and June 2020 in the University Hospital of Antwerp and between January 2017 and December 2019 in the Monzino Hospital of Milan were screened retrospectively for inclusion. Inclusion criteria were first episode of AF with electrocardiogram documentation before 45 years of age and first or repeated PVI procedure during the study period. A control group of all consecutive patients with first episode of ECG documented AF ≥45 years of age undergoing PVI between January 2016 and December 2019 in the University Hospital of Antwerp and between January 2017 and December 2017 in the Monzino Hospital of Milan was chosen. The study flow chart is shown in Supplementary Figure 1 . The definitions for the etiologic factors are described in the Supplementary Data . In both centers, participants provided informed consent to participate in the study. Young-AF was defined as having the first documented AF episode before the age of 45 years. Structural heart disease (SHD) was defined as the presence of a history of cardiac surgery, cardiomyopathy, or grown-up congenital heart disease. Paroxysmal supraventricular tachycardia (PSVT) was diagnosed if inducible during electrophysiologic study. Exercise level was defined as 0 (sedentary lifestyle), 1 (nonintensive, <2,000 hours of lifetime cumulative high-intensity exercise), or 2 (intensive, ≥2,000 hours of lifetime cumulative high-intensity exercise). Average alcohol consumption was assessed in units per week and the categories of alcohol consumption were defined as 0 (lifelong abstainer), 1 (mild, <7 units/week), 2 (moderate, 7 to 21 units/week), or 3 (heavy, >21 units/week). Familial AF was defined as a history of AF <60 years in ≥1 first-degree family member.
The choice between radiofrequency (RF) or cryoballoon (CB) ablation technique was left at the operator’s discretion and was based on availability. In all patients, complete electrophysiologic study was performed before or at the time of PVI. If PSVT was inducible, first ablation of the PSVT substrate was performed. A detailed description of the RF and CB PVI ablation techniques is provided in the Supplementary Data . Patients were evaluated with outpatient visits and ambulatory 24-hour Holter monitoring at 3, 6, and 12 months during the first year, and thereafter, every 6 to 12 months or earlier if symptoms developed. The study population of patients with young-AF was asked to provide extra information on risk factors using a questionnaire. A postprocedural blanking period of 90 days was considered. However, if recurrence of AF/atrial flutter/atrial tachycardia (AF/AFl/AT) during this blanking period resulted in a redo procedure, the patient was classified into the recurrence group. Patients were advised to stop antiarrhythmic drugs (AADs) after the blanking period. The Shapiro-Wilk test was used to examine if a continuous variable was normally distributed. Continuous variables with normal distribution were expressed as mean ± SD. Continuous variables without normal distribution were expressed as median (interquartile range). Categoric variables were expressed as percentages. The 2-sided Student t test for independent samples was used to compare continuous variables when values were homoscedastic, and its nonparametric equivalent (Mann-Whitney U test) was used when variables were heteroscedastic. The chi-square or Fisher’s exact test was used to compare categoric variables. Kaplan-Meier survival analysis was performed to analyze the cumulative event rates, and the log-rank test was used to detect significant differences between groups. To assess the contribution of baseline patient characteristics to recurrence multivariable Cox proportional hazard regression analysis was used. Only variables which were statistically significant in univariable analysis were used for the multivariable analysis. A p value of <0.05 was considered statistically significant. IBM SPSS Statistics (version 22; SPSS Inc., Chicago, Illinois) was used for all statistical analyses.
Comparison of baseline clinical and index procedural characteristics between the study population and control group are depicted in Table 2. In the University Hospital of Antwerp, 1 patient underwent CB and 433 patients underwent RF PVI. In the Monzino Hospital of Milan, 52/84 patients of the study population (62%) and 99/233 patients of the control group (42%) underwent CB PVI. A total of 8 patients with young-AF had an additional diagnosis of PSVT, 3 patients underwent successful PSVT ablation at the index PVI, and 5 patients had a history of PSVT ablation. Four patients with young-AF underwent a successful redo procedure during the blanking period of the index ablation owing to incessant AFl. These early events were counted as recurrence, but were not included in the Kaplan-Meier analysis. A total of 2 patients with young-AF were lost to follow-up. In the young-AF population, after a mean follow-up time of 36.1 ± 24.5 months, freedom from recurrent AF/AFl/AT (25% on AADs at last follow-up) was achieved in 138 patients (70%) after single PVI. Among the 59 patients with recurrence, repeated ablation was performed in 35 patients (at 13.7 ± 18.9 months). The Kaplan-Meier analysis revealed after 1 ablation in the young population 24% and 37% 1- and 4-year recurrence rate compared with 20% and 56% in the older population, respectively (p = 0.004) ( Figure 1) . After multiple ablations in the young population, 11% and 22% 1- and 4-year recurrence rate was observed compared with 16% and 45% in the older population, respectively (p <0.001) ( Figure 1 ). The estimated 5-year recurrence rate was 56% after 1 ablation and 34% after multiple ablations in the study population.
The baseline clinical and first procedural characteristics of the 195 patients with young-AF with follow-up are presented in Table 3 . In multivariable Cox proportional hazards regression, SHD was the only factor significantly associated with recurrence ( Table 4 ). However, the presence of more than 1 risk factor was associated with higher recurrence rate ( Figure 2 ). Each additional risk factor conferred 50% increased risk or recurrence (hazard ratio for each additional risk factor: 1.45; 95% confidence interval 1.1 to 1.8; p = 0.002). Comparison of baseline clinical and index procedural characteristics and outcome between patients undergoing RF and CB PVI in young-AF are shown in Table 5 . The CB group had a lower rate of AF/AFl/AT recurrence (37% in RF vs 11% in CB group, p = 0.001) and less redo procedures (23% in RF vs 6% in CB group, p = 0.006). After exclusion of patients with persistent AF and/or SHD (43/144 and 8/53 patients in the RF and CB group, respectively), Kaplan-Meier analysis did not show a significant difference between the RF and CB group (3-year recurrence: 27% in RF vs 17% in CB group, p = 0.11) ( Figure 3 ).
De With et al reported previously, in the largest cohort of relatively young (<60 years) patients with AF, that the large majority have AF in the setting of risk factors and 25% has a familial history of AF. In the present study, we investigated risk factors in patients undergoing PVI with an AF diagnosis at a much younger age (<45 years). The clinical outcomes of 7 previous small studies of PVI in patients with young-AF are summarized in Table 1 . These studies differed in definition of young-AF and ablation techniques and did not report in detail the diverse AF risk factors. To the best of our knowledge, our study represents the largest cohort of patients analyzing AF etiologic factors and clinical outcomes in patients with young-AF undergoing PVI with modern ablation technologies.
| First author | Country | Year | Study type | N | Inclusion criteria | Control group | Follow- up (months) | Main findings |
|---|---|---|---|---|---|---|---|---|
| Leong- Sit (4) | USA | 2010 | Single-center 100% RF | 232 | <45 y at initial procedure | Yes | Mean 32 |
|
| Chun (5) | Germany | 2013 | Multi-center, 80% RF | 593 | <45 y at initial procedure | Yes | 12 |
|
| Saguner (6) | Germany | 2018 | Single-center, 89% RF & 11% CB | 85 | <35 y at initial procedure | No | Median 54 |
|
| Moran (7) | Belgium | 2018 | Single-center, 100% CB | 57 | <40 y at initial procedure | No | Median 18 |
|
| Chen (8) | China | 2019 | Single-center, 100% RF | 75 | <35 y at initial procedure | No | Median 61 |
|
| Allam (9) | Egypt | 2019 | Single-center, 100% RF | 34 | <35 y at initial procedure | Yes | 12 |
|
| Bergau (10) | Germany | 2020 | Single-center, 100% CB | 93 | <45 y at initial procedure | Yes | Mean 32 |
|
| Total | <45 years | ≥45 years | p Value | |||||
|---|---|---|---|---|---|---|---|---|
| Number of patients | 751 | 197 | 554 | |||||
| Baseline clinical characteristics | ||||||||
| Age at diagnosis of AF (years) | 53,4 ± 14,0 | 36,2 ± 7,0 | 61,3 ± 8,1 | <0,001 | ||||
| Age at index ablation (years) | 57,9 ± 13,2 | 40,5 ± 8,1 | 64,1 ± 8,1 | <0,001 | ||||
| Paroxysmal AF | 489 (65%) | 160 (81%) | 329 (60%) | <0,001 | ||||
| Men | 525 (70%) | 162 (82%) | 363 (66%) | <0,001 | ||||
| BMI (kg/m 2 ) Mean >30 | 26,8 ± 4,4 131 (17%) | 26,3 ± 4,7 38 (19%) | 27,0 ± 4,3 129 (23%) | 0,054 0,6 | ||||
| Height (cm) | 175 ± 9,7 | 179,9 ± 9,4 | 173 ± 9,3 | <0,001 | ||||
| Weight (kg) | 82,5 ± 15,8 | 85,6 ± 18,4 | 81,4 ± 14,6 | 0,005 | ||||
| Body surface area (m 2 ) | 1,9 ± 0,4 | 1,98 ± 0,3 | 1,85 ± 0,4 | <0,001 | ||||
| Heart failure | 112 (15%) | 17 (9%) | 95 (17%) | 0,003 | ||||
| Hypertension | 295 (39%) | 42 (21%) | 253 (46%) | <0,001 | ||||
| Diabetes mellitus | 61 (8%) | 4 (2%) | 57 (10%) | <0,001 | ||||
| TIA/Stroke | 56 (8%) | 7 (4%) | 49 (9%) | <0,001 | ||||
| Coronary artery disease | 91 (12%) | 11 (6%) | 80 (14%) | 0,001 | ||||
| CHA2DS2-VASc score | 1,5 ± 1,4 | 0,54 ± 0,8 | 1,9 ± 1,4 | <0,001 | ||||
| Obstructive sleep apnea syndrome | 50/518 (10%) | 17 (9%) | 33/321 (10%) | 0,6 | ||||
| Chronic kidney disease | 6/395 (2%) | 0 (0%) | 6/198 (3%) | 0,03 | ||||
| Previous cardiac surgery | 26/395 (7%) | 4 (2%) | 19/198 (10%) | 0,02 | ||||
| Tachycardia-induced cardiomyopathy | 37/392 (9%) | 13 (7%) | 24/195 (12%) | 0,06 | ||||
| Left atrial diameter (mm) | 42,6 ± 7,0 | 40,9 ± 6,2 | 44,2 ± 7,2 | <0,001 | ||||
| Left atrial volume index (ml/m 2 ) | 38,4 ± 12,7 | 33,2 ± 10,9 | 40,2 ± 12,7 | <0,001 | ||||
| Left ventricular ejection fraction (%) | 58,9 ± 10,9 | 60,3 ± 8,6 | 58,5 ± 11,2 | 0,3 | ||||
| Antiarrhythmic drugs Flecainide, Sotalol Amiodarone | 192/448 (43%) 94/448 (21%) 82/448 (18%) | 121 (61%) 33 (17%) 18 (9%) | 71/251 (28%) 61/251 (24%) 64/251 (26%) | <0,001 0,06 <0,001 | ||||
| Cardiac implantable electronic device | 46/518 (9%) | 12/203 (6%) | 34/321 (10%) | 0,1 | ||||
| Previous PVI | 39 (5%) | 12 (6%) | 27 (5%) | 0,5 | ||||
| Index PVI procedure | ||||||||
| RF | 599 (79,8%) | 144 (73%) | 455 (82%) | 0,01 | ||||
| Sinus rhythm at begin of procedure | 360/516 (70%) | 160 (82%) | 200/321 (62%) | <0,001 | ||||
| First pass isolation * | 247/405 (61%) | 63/102 (62%) | 184/303 (61%) | 0,9 | ||||
| Fluoroscopy time (min) | 18,2 ± 11 | 17,3 ± 9 | 18,4 ± 11,8 | 0,4 | ||||
| Complications | ||||||||
| Any major complications | 26/446 (6%) | 8 (4%) | 18/249 (7%) | 0,2 | ||||
| Tamponade | 3 (1%) | 0 (0%) | 3 (1%) | |||||
| Acute coronary syndrome | 1 (0%) | 0 (0%) | 1 (0,5%) | |||||
| Sinus arrests leading to pacemaker Implantation | 1 (0%) | 0 (0%) | 1 (0,5%) | |||||
| Arteriovenous fistula | 6 (1%) | 4 (2%) | 2 (1%) | |||||
| Pseudoaneurysm | 8 (2%) | 2 (1%) | 6 (2%) | |||||
| Hematoma for which intervention was Warranted | 2 (0,5%) | 1 (0,5%) | 1 (0,5%) | |||||
| Deep-vein thrombosis | 2 (0,5%) | 1 (0,5%) | 1 (0,5%) | |||||
| Stroke | 3 (1%) | 0 (0%) | 3 (1%) | |||||
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
