Burden and Long Firing of Premature Atrial Contraction Early After Catheter Ablation Predict Late Recurrence of Atrial Fibrillation

Abstract

Background: Associations between characteristics of premature atrial contraction (PAC) 6 months after catheter ablation (CA) and later recurrence are not known. We investigated the effects of PAC characteristics on long-term outcomes of initially successful atrial fibrillation (AF) ablation.

Methods and Results: In all, 378 patients (mean age 61 years, 21% female, 67% paroxysmal AF) who underwent initial radiofrequency CA for AF without recurrence up to 24-h Holter monitoring 6 months after the procedure were reviewed retrospectively. The calculated number of PAC/24 h and the length of the longest PAC run during Holter recording were analyzed. After 4.3±1.2 years (mean±SD) follow-up, 123 (32.5%) patients experienced late recurrence. Patients with recurrence had significantly more PAC/24 h (median [interquartile range] 110 [33–228] vs. 42 [16–210]; P<0.01) and a longer longest PAC run (5 [2–8] vs. 3 [1–5]; P<0.01) than those without. Receiver operating characteristic curve analysis indicated 58 PAC/24 h and a longest PAC run of 5 were optimal cut-off values for predicting recurrence. After adjusting for previously reported predictors of late recurrence, frequent PAC (≥58/24 h) and longest PAC run ≥5 were found to be independent predictors of late recurrence (hazard ratios [95% confidence intervals] 1.93 [1.24–3.02; P<0.01] and 1.81 [1.20–2.76; P<0.01], respectively).

Conclusions: Six months after successful AF ablation, both frequent PAC and long PAC run are independent predictors of late recurrence.

Since Haïssaguerre et al revealed that the origin of premature atrial contractions (PACs), which initiated atrial fibrillation (AF), was mostly from the pulmonary veins (PVs),¹ catheter ablation (CA) has become a widely accepted intervention for rhythm control of AF. The current standard strategy of CA for AF is PV isolation (PVI).

Editorial p 883

It is known that a certain number of initially successful cases soon after CA experience late recurrence, even years after the procedure, and it is very common to find recovered PV conduction in such cases.² Detecting late recurrence is beneficial to improving patients’ prognosis and quality of life, because this information can help in the selection of antiarrhythmic drugs, determining the necessity for repeat sessions, and managing heart failure. Previous studies reported that persistent AF, age, hypertension, dyslipidemia, wide area circumferential ablation, weight >90.72 kg, presence of non-PV triggers, C-reactive protein >0.5 mg/dL, reduction of left atrial (LA) volume, and isolation of fewer PVs were independent factors associated with late recurrence.^3–8

With regard to findings during Holter monitoring after CA, two previous reports indicated that the number of PACs at 6 months⁹ and 2–8 days¹⁰ after the procedure was a predictor of future recurrence. However, this association has not been established and the underlying mechanism has not yet been determined. Furthermore, PAC characteristics, other than the number of PAC per day, were not examined in those studies. Hence, the aim of the present study was to clarify the effects of PAC characteristics, evaluated on the basis of Holter monitoring findings, on the long-term outcome of initially successful AF ablation.

Methods

Study Design and Population

This study was a single-center retrospective observational study that enrolled consecutive patients who underwent radiofrequency (RF) CA for AF at Sakurabashi Watanabe Hospital between March 2012 and March 2014. All patients underwent PVI by RF catheter, and some underwent additional ablation if necessary. Patients were excluded if they experienced recurrence before the scheduled 24-h Holter monitoring at 6 months after the procedure with a 3-month blanking period or if they did not undergo Holter monitoring at 6 months after CA. Thus, only patients who did not have AF recurrence before the end of the 24-h Holter monitoring were investigated (Figure 1). Clinical data were collected from patients’ medical records.

Figure 1.

Flowchart of patient inclusion. AF, atrial fibrillation; RFCA, radiofrequency catheter ablation.

This study was approved by the Sakurabashi Watanabe Hospital Ethics Committee (Reference no. 18-48). All patients provided written informed consent for the ablation procedure and the retrospective use of their clinical data.

Ablation Procedure

The detailed electrophysiological studies and ablation protocol have been described elsewhere.¹¹ Patient respiration was controlled by bilevel positive airway pressure. A 6-Fr decapolar catheter was placed in the coronary sinus via the median antebrachial vein and a 7-Fr decapolar catheter was placed in the superior vena cava and right atrium. A 10-Fr SoundStar ultrasound catheter (Biosense Webster, Diamond Bar, CA, USA) was inserted into the right atrium via the femoral vein for the anatomic mapping of the LA using the CartoSound module of the CARTO3 system (Biosense Webster). Two 8-Fr long sheaths (Daig SL0; St Jude Medical, St Paul, MN, USA) were introduced into the LA via the femoral vein after a single Brockenbrough puncture using a RF needle (Japan Lifeline, Tokyo, Japan). An initial intravenous bolus of heparin (150 IU/kg) was followed by continuous infusion (300 IU/kg/h) to maintain an activated clotting time >300 s. RF ablation was conducted using the ThermoCool SmartTouch (Biosense Webster). Electrical isolation of PVs was assessed by a 20-polar ring catheter (LASSO; Biosense Webster). The endpoint of the PVI was the achievement of a bidirectional conduction block between the LA and the PVs. After the completion of PVI, isoproterenol (4–20 µg/min) was infused to examine induced PV reconnections and/or non-PV AF triggers. Ablation of non-PV AF triggers was strongly encouraged. Other additional ablations, such as linear ablation and complex fractionated atrial electrogram ablation, depended on the operator’s judgment. The procedures were performed with patients under conscious sedation with pentazocine, thiamylal sodium, and dexmedetomidine hydroxyzine. All antiarrhythmic drugs, including β-blockers, were discontinued for at least 5 half-lives before the ablation. The ablation protocol for second procedures was the same as for first procedure. We searched for reconnection of PV potentials, identified AF triggers, including those derived from non-PV origins, and attempted to ablate them.

Patient Follow-up and Study Endpoints

Patients were followed-up at 1, 3, and 6 months after the ablation and were instructed to undergo electrocardiography (ECG) in case of symptoms suggestive of arrhythmia recurrence. Patients were encouraged to check their pulse rate and rhythm three times a day and to visit Sakurabashi Watanabe Hospital if they experienced irregular pulses. In such cases, inspections were conducted using an external loop recorder. Patients routinely undergo 24-h Holter monitoring (RAC-3203 and DSC-5500; Nihon Kohden, Tokyo, Japan) at the 6-month follow-up. After Holter monitoring 6 months after the procedure, scheduled ECG follow-up was performed at least every 6 months. When patients noticed an irregularity of their pulse or symptoms suggestive of recurrence, they were encouraged to visit Sakurabashi Watanabe Hospital. ECG was also performed at every additional visit, and additional Holter ECG monitoring was performed and/or event monitor recordings obtained, if required. Patients who underwent a second ablation procedure also underwent 24-h Holter monitoring and were followed-up using the same schedule as that used initially. Holter monitoring data were analyzed by experienced cardiac technicians and reviewed by the attending cardiologists. PAC was defined as a supraventricular complex occurring >30% earlier than the expected R-R interval. Given that there was a variation in the duration of Holter monitoring among patients, PAC/24 h was defined as the calculated number of PAC within a 24-h period. In addition, the longest PAC run was defined as the number of beats in the longest PAC run during Holter monitoring. AF recurrence was defined as documented AF or atrial tachyarrhythmia >30 s. Antiarrhythmic drugs were defined as Vaughan Williams Class I and III drugs and bepridil. Discontinuation of antiarrhythmic drugs was strongly encouraged after the blanking period. As described above, patients who experienced AF recurrence before the scheduled 24-h Holter monitoring at 6 months after the procedure with a 3-month blanking period were excluded from the analysis. The primary end point was late recurrence of AF, which was defined as initial AF recurrence after the Holter monitoring at 6 months after the index procedure.

Statistical Analysis

Parametric and non-parametric variables are given as the mean±SD and the median and interquartile range (IQR), respectively. Given that the distribution of C-reactive protein, PAC/24 h, and longest PAC run was non-parametric, these data were analyzed using the Wilcoxon rank-sum test. The significance of differences in patient characteristics between those with and without late recurrence was assessed using Student’s t-test or the Chi-squared test. Receiver operating characteristic (ROC) curve analysis was used to determine the optimal cut-off value of PAC/24 h and longest PAC run to predict late recurrence. These cut-off values were then used to stratify the patients. The time to AF recurrence from ablation was illustrated using the Kaplan-Meier method, and the significance of differences among the stratified population was assessed by log-rank test. Patients were censored on the day of the last follow-up. Univariate and multivariate hazard ratios (HRs) were determined by Cox proportional hazard regression models to assess the association of patient characteristics, echo parameter, and Holter parameters with late recurrence. Two-tailed P<0.05 was considered significant. HRs of PAC/24 h ≥58 and longest PAC run ≥5 for late recurrence were evaluated after adjusting for previously known risk factors of late recurrence, namely diabetes, age, sex, hypertension, non-paroxysmal AF, body mass index (BMI), non-PV triggers, and LA diameter. Statistical analyses were performed using JMP Pro 13.1. (SAS Institute, Cary, NC, USA).

Results

Patient Characteristics

Of the 575 consecutive patients, 197 patients who experienced recurrence before the scheduled 24-h Holter monitoring 6 months after the procedure or those who did not undergo Holter monitoring 6 months after CA were excluded. Thus, data from 378 patients were evaluated.

The mean patient age was 60.7±10.2 years. Of the patients in the study, 80 (21.2%) were women and 252 (66.7%) had paroxysmal AF. Significantly more patients with late recurrence had long-standing persistent AF. Details of the patients included in the study are given in Table 1. There were 204 (54.0%) patients with a CHA₂DS₂-VASc score of 0 or 1; the study population was composed of relatively low-risk patients. Results of echo parameter analysis showed that the patients had good left ventricular function and a small LA diameter. All patients underwent PVI, with 28.3% of these patients undergoing cavotricuspid isthmus (CTI) block line. However, few patients underwent additional ablation other than the CTI block line (Table 2). At the time of Holter monitoring, the prevalence of an antiarrhythmic drug prescription was higher among those with late recurrence (29.3% vs. 18.0%; P=0.01), whereas there was no significant different in β-blocker prescriptions (20.3% vs. 18.0%; P=0.59; Table 3).

Table 1. Patient Characteristics

	All patients (n=378)	AF recurrence		P-value
		Yes (n=123)	No (n=255)
Age (years)	60.7±10.2	61.5±10.5	60.3±10.1	0.32
Female sex	80 (21.1)	33 (26.9)	47 (18.4)	0.06
Height (cm)	167.4±9.0	166.4±9.3	167.8±8.8	0.15
Weight (kg)	68.2±12.7	67.1±12.6	68.8±12.7	0.23
BMI (kg/m2)	24.2±3.4	24.1±3.5	24.3±3.4	0.64
Congestive heart failure	43 (11.3)	16 (13.0)	27 (10.6)	0.49
Hypertension	177 (46.8)	61 (50.0)	116 (45.4)	0.45
Diabetes	45 (11.9)	9 (7.3)	36 (14.1)	0.06
Ischemic stroke	26 (6.9)	9 (7.3)	17 (6.7)	0.81
CHA2DS2-VASc score
0 or 1	204 (54.0)	63 (51.2)	141 (55.3)	0.46
2 or 3	141 (37.3)	50 (40.7)	91 (35.7)	0.35
≥4	33 (8.7)	10 (8.1)	23 (9.0)	0.80
Paroxysmal AF	252 (66.7)	80 (65.0)	172 (67.5)	0.59
Persistent AF	86 (22.8)	24 (20.0)	62 (24.3)	0.30
Long-standing AF	40 (10.6)	19 (15.4)	21 (8.2)	0.03
Hemoglobin (g/dL)	14.4±1.3	14.2±1.4	14.6±1.2	0.02
Creatinine (mg/dL)	0.83±0.20	0.83±0.22	0.83±0.19	0.80
HbA1c (%)	5.8±0.6	5.8±0.6	5.8±0.5	0.86
Uremic acid (mg/dL)	5.8±1.4	5.9±1.4	5.8±1.4	0.36
Total cholesterol (mg/dL)	205.6±37.5	207.4±35.9	204.8±38.3	0.53
LDL-C (mg/dL)	120.3±31.2	122.3±29.5	119.4±32.0	0.40
HDL-C (mg/dL)	57.6±15.2	57.0±14.4	57.8±15.6	0.61
Triglyceride (mg/dL)	122.4±70.3	123.8±76.4	121.8±67.3	0.80
CRP (mg/dL)	0.05 [0.03–0.1]	0.05 [0.03–0.1]	0.05 [0.03–0.1]	0.71
IVSd (mm)	10.3±1.8	10.4±1.8	10.2±1.8	0.48
PWd (mm)	10.0±1.3	10.2±1.4	9.9±1.3	0.04
E/A ratio	0.97±0.31	0.96±0.37	0.97±0.29	0.80
Mitral E/e’ ratio	9.9±4.7	10.2±4.8	9.7±4.7	0.41
LVEF (%)	65.1±8.9	65.5±9.0	64.9±8.8	0.48
LVEDD (mm)	46.1±5.0	46.3±4.7	46.1±5.2	0.68
LA diameter (mm)	37.3±5.9	37.9±5.9	37.0±5.9	0.17
LA volume (mL)	47.3±21.3	45.2±16.0	48.3±23.5	0.24
LA volume index (mL/m2)	28.1±13.4	26.4±9.5	29.0±15.0	0.11

Parametric continuous variables are given as the mean±SD; non-parametric continuous variables are given as the median [interquartile range]. Categorical variables are given as n (%). A, late diastolic mitral valve inflow velocity; AF, atrial fibrillation; BMI, body mass index; CRP, C-reactive protein; E, early diastolic mitral valve inflow velocity; e’, early diastolic mitral annular velocity; HDL-C, high-density lipoprotein cholesterol; IVSd, diastolic interventricular septum wall thickness; LDL-C, low-density lipoprotein cholesterol; LA, left atrial; LVEDD, left ventricular end-diastolic diameter; LVEF, left ventricular ejection fraction; PWd, diastolic posterior wall thickness.

Table 2. Details of the Ablation Procedure (n=378)

PV isolation	378 (100)
CTI block line	107 (28.3)
Roof line of left atrium	4 (1.1)
Mitral valve isthmus block line	3 (0.8)
SVC isolation	6 (1.6)
Non-PV foci ablation	5 (1.3)
CFAE ablation	4 (1.1)
GP ablation	0 (0)

Data are given as n (%). CFAE, complex fractionated atrial electrogram; CTI, cavotricuspid isthmus; GP, ganglionated plexi; PV, pulmonary vein; SVC, superior vena cava.

Table 3. Details of Antiarrhythmic Medications at the Time of Holter Monitoring

	AF recurrence		P-value
	Yes (n=123)	No (n=255)
β-blocker	25 (20.3)	46 (18.0)	0.59
All antiarrhythmic drugsA	36 (29.3)	46 (18.0)	0.01
Class I	20 (16.3)	28 (11.0)	0.15
Class III	2 (0.78)	0 (0)	0.32
Bepridil	19 (15.5)	14 (5.5)	<0.01

Data are given as n (%). ^AAll antiarrhythmic drugs includes Vaughan Williams Class I and III drugs and bepridil. AF, atrial fibrillation.

Late Recurrence and Holter Monitoring

During a mean follow-up period of 4.3±1.2 years, 123 (32.5%) patients experienced AF recurrence. Holter monitoring detected a median of 53 (IQR 18–322) PACs over a period of 22.6±1.1 h, with PAC/24 h calculated as 57 (IQR 19–341); the median longest PAC run was 3 (IQR 2–6). Patients with recurrence had significantly more PAC/24 h and a greater longest PAC run than those without recurrence (108 [33, 996] vs. 42 [16, 210] PAC/24 h, respectively [P<0.01]; longest PAC run 5 [2, 8] vs. 3 [1, 5], respectively [P<0.01]). Other Holter monitoring data are provided in Table 4.

Table 4. Holter Monitoring Data

	All patients (n=378)	AF recurrence		P-value
		Yes (n=123)	No (n=255)
Monitoring duration (h)	22.6±1.1	22.5±1.2	22.6±1.0	0.62
Total no. heart beats/24 h	106,784.4±13,979.8	105,839.5±15,628.0	107,240.0±13,120.0	0.39
Heart rate (beats/min)
Maximum	116.9±18.0	115.2±17.4	117.7±18.3	0.21
Minimum	53.9±8.8	52.6±10.3	54.5±7.9	0.05
Mean	75.5±9.8	74.8±11.0	75.9±9.2	0.33
No. PAC/24 h (beats)	57 [19–341]	108 [33–996]	42 [16–210]	<0.01
PAC burden (%)	0.05 [0.017–0.32]	0.12 [0.03–0.96]	0.04 [0.015–0.19]	<0.01
Longest PAC run (beats)	3 [2–6]	5 [2–8]	3 [1–5]	<0.01

Parametric continuous variables are given as the mean±SD; non-parametric continuous variables are given as the median [interquartile range]. AF, atrial fibrillation; PAC, premature atrial contraction.

ROC curve analysis revealed that optimal cut-off values of 58 PAC/24 h and a longest PAC run of 5 could predict recurrence with a C-statistic of 0.63 (95% confidence interval [CI] 0.57–0.69) and 0.62 (95% CI 0.56–0.68), respectively (Figure 2). The Kaplan-Meier plot of AF recurrence showed that the AF recurrence rate was significantly higher in patients with PAC/24 h ≥58 than in those with PAC/24 h <58 (log-rank test, P<0.01). Patients with a longest PAC run ≥5 also had a significantly higher AF recurrence rate (log-rank test, P<0.01; Figure 3A,B). After classifying patients into 4 groups using the cut-off values of 58 PAC/24 h and a longest PAC run of 5, we found that patients with PAC/24 h ≥58 and a longest PAC run ≥5 had the highest recurrence rate. Conversely, patients with PAC/24 h <58 and a longest PAC run <5 had the lowest recurrence rate, and those with either a PAC/24 h ≥58 or a longest PAC run ≥5 had an intermediate risk of recurrence (Figure 3C). The late recurrence rate in patients with both, either, or neither PAC/24 h ≥58 and longest PAC run ≥5 was 57.3% (95% CI 46.0–68.0%), 44.2% (31.3–58.0%), and 21.8% (15.4–29.9%), respectively, which can be classified as high, intermediate, and low risks. The HR of late recurrence among the high- and low-risk groups was 3.29 (95% CI 2.13–5.17; P<0.01).

Figure 2.

Receiver operating characteristic curves of (A) premature atrial contraction (PAC)/24 h and (B) longest PAC run for late atrial fibrillation recurrence. The C-statistics for PAC/24 h and longest PAC run were 0.63 (95% confidence interval [CI] 0.57–0.69) and 0.62 (95% CI 0.56–0.68), respectively.

Figure 3.

Risk of late atrial fibrillation (AF) recurrence and premature atrial contraction (PAC). Kaplan-Meier plots show the late AF recurrence rate stratified by (A) the calculated number of PAC per 24 h, (B) longest PAC run (during Holter recording), and (C) both.

Adjusted Risk for Late Recurrence

The univariate Cox proportional hazard regression model showed that PAC/24 h ≥58 and a longest PAC run ≥5 were significantly associated with a high late recurrence rate. There was no significant correlation between PAC/24 h and longest PAC run (r=0.087, P=0.09); thus, we included both in the multivariate Cox proportional hazard regression analysis adjusted for previously reported risk factors of late recurrence. The adjusted HRs of PAC/24 h ≥58 and a longest PAC run ≥5 for late recurrence were 1.93 (95% CI 1.24–3.02; P<0.01) and 1.81 (1.20–2.76; P<0.01), respectively (Table 5).

Table 5. Cox Proportional Hazards Regression Model of Patient Characteristics, Ablation Procedure, and Holter Monitoring Variables Associated With AF Recurrence

	Univariate		Multivariate
	HR (95% CI)	P-value	HR (95% CI)	P-value
Age	1.01 (0.99–1.03)	0.47	0.98 (0.96–1.01)	0.15
Male	0.73 (0.49–1.10)	0.13	0.83 (0.54–1.30)	0.41
BMI	0.99 (0.94–1.05)	0.8	1.03 (0.93–1.07)	0.54
Hypertension	1.14 (0.80–1.62)	0.48	1.19 (0.81–1.74)	0.37
Diabetes	0.53 (0.25–0.98)	0.04	0.5 (0.23–0.95)	0.03
Non-PAF	1.1 (0.75–1.59)	0.61	1.01 (0.67–1.50)	0.98
Non-PV trigger	0.68 (0.31–1.31)	0.29	1.42 (0.55–3.03)	0.44
LAD ≥42 mm	1.2 (0.77–1.79)	0.41	1.06 (0.66–1.66)	0.8
PAC/24 h ≥58	2.38 (1.65–3.49)	<0.01	1.93 (1.24–3.02)	<0.01
Longest PAC run ≥5	2.34 (1.64–3.35)	<0.01	1.81 (1.20–2.76)	<0.01

CI, confidence interval; HR, hazard ratio; LAD, left atrial diameter; PAF, paroxysmal atrial fibrillation. Other abbreviations as in Tables 1,2,4.

Findings After the Second Procedure

In all, 71 patients underwent a second procedure due to AF recurrence. The strategy for the second procedure was to search for reconnected PV potentials, provoked non-PV AF triggers, and attempted to eliminate them; 24-h Holter monitoring was also performed 6 months after the second procedure. According to results of electrophysiological studies after the second procedure, most patients (n=65; 92%) showed reconnection of PVs. Among the 71 patients who underwent a second procedure, 33 received re-PVI alone. In these 33 patients, PAC/24 h decreased significantly after the second procedure (from a median of 239 [IQR 37–1,567] to 101 [IQR 27–339]; P<0.01). However, the longest PAC run did not decrease after the second procedure in these patients (from a median of 5 [IQR 3–8] to 4 [IQR 2–7]; P=0.29).

Discussion

Main Findings

The findings of this study indicate that the PAC burden and PAC firing observed at 6 months after CA are predictors of late AF recurrence over the long-term follow-up. The novel findings of this study are that not only is PAC burden an independent predictor of late recurrence, but so too is the length of the longest PAC run; the combination of these two factors allowed the stratification of patients into high-, intermediate-, and low-risk groups. Furthermore, the finding that more PACs were observed in patients with than without late recurrence was due primarily to the reconnected PVs.

Comparison With Previous Studies

Previous studies have reported an association between PAC burden and AF recurrence,^9,10,12 which is consistent with the findings of the present study. In particular, Gang et al⁹ investigated the association between AF recurrence and PAC burden 6 months after the procedure, a timing similar to that of the present study. However, the PAC/24 h cut-off value determined in the present study (58) was less than that reported by Gang et al⁹ (142). We speculate that this could be due primarily to the patients in the present study having less advanced atrial remodeling; the mean LA diameter of the cohort in this study was 37.3 mm, which is less that that reported in previous studies (mean LA diameter ranging between 42.5 and 44.3 mm).¹³ However, patient age, sex, and AF subtype were similar between the present study and that of Gang et al.¹³ We also speculate that the differences could also be due, in part, to the present cohort including more patients who were using antiarrhythmic drugs at the time of Holter monitoring: the prescription rate for antiarrhythmic drugs in the present study (21.7%) is slightly higher than that reported by Gang et al⁹ (between 14% and 18%). Specifically in the present cohort, patients with late recurrence had more frequent PACs and a longer PAC run, despite the higher prevalence of antiarrhythmic medication, than those without recurrence. This is probably because the attending cardiologists hesitated to discontinue antiarrhythmic medication in cases with frequent PACs early after the procedure.

In addition to PAC/24 h, we evaluated the association between the length of the PAC run and late recurrence and found that a longer PAC run was significantly associated with late recurrence. Because the PAC burden and the length of the PAC run were not significantly correlated, we used both these factors for risk stratification for late AF recurrence. Consequently, the combination of PAC/24 h and longest PAC run allowed further risk stratification of patients into high-, intermediate-, and low-risk groups for late recurrence (Figure 3C). Previous studies did not investigate the mechanism of association between the increased number of PACs and late recurrence. To the best of our knowledge, the present study is the first to provide evidence supporting PACs originating from reconnected PVs as the mechanism contributing to the significant association between the increased number of PACs and late recurrence, as discussed below.

Origin of PACs

We evaluated electrophysiological findings for patients with late recurrence undergoing a second procedure and their follow-up Holter monitoring data. We focused on patients who underwent successful re-PVI alone and had no additional ablation in the second procedure because ablation of non-PV arrhythmogenicity may affect PACs from non-PV foci. A significant decrease in the PAC burden in these patients indicated that the main site of origin of the PACs was the reconnected PVs. In contrast, the length of the longest PAC run did not change after the second procedure in which the PV potential was successfully re-eliminated. These data indicate that the longest PAC runs were repetitive firings originating from non-PV foci, or “short AF” sustained for <30 s and initiated due to a non-PV arrhythmogenic substrate. Those with a longest PAC run ≥5 are older and more likely to be female than those with a longest PAC run <5 (age 64.5±8.9 vs. 58.6±10.3 years, respectively [P<0.01]; females 30.4% vs. 15.8%, respectively [P<0.01]), although the LA diameter (37.9±6.0 vs. 37.0±5.9 mm; P=0.13) and the prevalence of non-paroxysmal AF (30.4% vs 34.6%; P=0.40) did not differ significantly between the 2 group. Given that age and female sex have been reported as factors for the high prevalence of a low voltage area in the LA,¹⁴ we hypothesize that the length of the PAC run reflects non-PV arrhythmogenicity.

Mechanism of Late Recurrence Development

Our data indicated that the number of PACs and the length of the PAC run were independently associated with the risk of late recurrence. The increased number of PACs in patients with late recurrence was due primarily to the reconnected PVs. PV reconnection is presumed to occur early on after the PVI procedure. In this study we electively included patients without recurrence at 6 months after the index procedure, and even those with PV reconnection did not experience recurrence for 6 months despite the PV reconnection, which suggests that these patients did not have a suitable substrate to initiate AF at that time point. With regard to the length of the PAC run, the long PAC firings may include “short AF”, which is not sustained for more than 30s. Non-sustained AF may develop due to the absence of sufficient substrate required to perpetuate AF. Because arrhythmogenic substrates can progress with time,¹⁵ frequent PACs and long PAC runs may be suitable triggers to initiate and perpetuate AF during the follow-up period owing to the newly developed substrate, which would result in late recurrence. Therefore, the management of risk factors associated with the progression of AF substrate, such as hypertension, sleep apnea, heart failure, and obesity, would be especially important in these patients.

Clinical Implications

Considering that both the number of PACs and the length of PAC firing are relatively easy to assess through 24-h Holter monitoring, these parameters are useful in clinical practice to identify patients who are at high risk of AF recurrence and need careful observation. It has been reported that AF ablation possibly reduces the risk of stroke and mortality^16–19 and AF recurrence was associated with an increased stroke rate.¹⁷ Thus, early recognition of AF recurrence is expected to be useful in reducing the risk of stroke and heart failure.

Study Limitations

The study findings should be interpreted with consideration of the study limitations. First, given that this study was observational in nature and conducted at a single center, the results are affected by the patients’ background and the ablation strategy used at the center. Hence, the cut-off values for PAC/24 h and longest PAC run cannot be applied universally. In particular, the present study cohort had a small LA diameter, indicating that these patients are in the less advanced stage of substrate progression than the general population undergoing CA for AF. Second, we do not know whether performing Holter monitoring at 6 months after CA is the best time to predict late recurrence, because we have no other Holter monitoring data. Third, we cannot ignore the possibility of underestimating the AF recurrence rate owing to asymptomatic recurrence. Fourth, the prevalence of the sleep apnea syndrome, which causes the progression of LA remodeling, is unknown because of the data were not available. Finally, we could only demonstrate the major origin of the PAC burden after CA in patients with late recurrence in some patients. The coupling interval of PACs can be an indication because PACs derived from the PVs have a short coupling interval, as reported previously.^20,21 However, no data were available regarding the coupling interval of PACs in the present study. Moreover, the resolution of the Holter recording was not suitable to investigate PAC origins based on P-wave morphology. Therefore, further studies are needed to elucidate the underlying mechanism.

Conclusions

The data indicate that PAC burden and PAC run in initially successful AF ablation cases are highly associated with late recurrence. An increase in PACs from reconnected PVs was suggested as the mechanism contributing to the significant association between the number of PACs and AF recurrence. It is clinically important to carefully monitor patients with high PAC burden or long PAC firing after an initially successful AF ablation.

Acknowledgement

The authors are grateful to all the staff at Sakurabashi Watanabe Hospital for helping conduct this study.

Sources of Funding

This research received no grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of Interest

K. Inoue has received honoraria from Johnson and Johnson KK, Medtronic Japan, Bayer Yakuhin, Nihon Boehringer Ingelheim, Bristol Myers Squibb, and Daiichi-Sankyo. The other authors declare no conflicts of interest.

Data Availability

Deidentified participant data will not be shared.

References

• 1.   Haïssaguerre M, Jaïs P, Shah DC, Takashi A, Hocini M, Quiniou G, et al. Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J Med 1998; 339: 659–666.
• 2.   Ouyang F, Tilz R, Chun J, Schmidt B, Wissner E, Zerm T, et al. Long-term results of catheter ablation in paroxysmal atrial fibrillation: Lessons from a 5-year follow-up. Circulation 2010; 122: 2368–2377.
• 3.   Sotomi Y, Inoue K, Ito N, Kimura R, Toyoshima Y, Masuda M, et al. Incidence and risk factors for very late recurrence of atrial fibrillation after radiofrequency catheter ablation. Europace 2013; 15: 1581–1586.
• 4.   Sotomi Y, Inoue K, Tanaka K, Toyoshima Y, Oka T, Tanaka N, et al. Persistent left atrial remodeling after catheter ablation for non-paroxysmal atrial fibrillation is associated with very late recurrence. J Cardiol 2015; 66: 370–376.
• 5.   Wokhlu A, Hodge DO, Monahan KH, Asirvatham SJ, Friedman PA, Munger TM, et al. Long-term outcome of atrial fibrillation ablation: Impact and predictors of very late recurrence. J Cardiovasc Electrophysiol 2010; 21: 1071–1078.
• 6.   Mainigi SK, Sauer WH, Cooper JM, Dixit S, Gerstenfeld EP, Callans DJ, et al. Incidence and predictors of very late recurrence of atrial fibrillation after ablation. J Cardiovasc Electrophysiol 2007; 18: 69–74.
• 7.   Shah AN, Mittal S, Sichrovsky TC, Cotiga D, Arshad A, Maleki K, et al. Long-term outcome following successful pulmonary vein isolation: Pattern and prediction of very late recurrence. J Cardiovasc Electrophysiol 2008; 19: 661–667.
• 8.   Seow SC, Lim TW, Koay CH, Ross DL, Thomas SP. Efficacy and late recurrences with wide electrical pulmonary vein isolation for persistent and permanent atrial fibrillation. Europace 2007; 9: 1129–1133.
• 9.   Gang UJ, Nalliah CJ, Lim TW, Thiagalingam A, Kovoor P, Ross DL, et al. Atrial ectopy predicts late recurrence of atrial fibrillation after pulmonary vein isolation. Circ Arrhythm Electrophysiol 2015; 8: 569–674.
• 10.   Alhede C, Johannessen A, Dixen U, Jensen JS, Raatikainen P, Hindricks G, et al. Higher burden of supraventricular ectopic complexes early after catheter ablation for atrial fibrillation is associated with increased risk of recurrent atrial fibrillation. Europace 2018; 20: 50–57.
• 11.   Oka T, Inoue K, Tanaka K, Ninomiya Y, Hirao Y, Tanaka N, et al. Left atrial reverse remodeling after catheter ablation of nonparoxysmal atrial fibrillation in patients with heart failure with reduced ejection fraction. Am J Cardiol 2018; 122: 89–96.
• 12.   Alhede C, Lauridsen TK, Johannessen A, Dixen U, Jensen JS, Raatikainen P, et al. The impact of supraventricular ectopic complexes in different age groups and risk of recurrent atrial fibrillation after antiarrhythmic medication or catheter ablation. Int J Cardiol 2018; 250: 122–127.
• 13.   Lim TW, Koay CH, See VA, McCall R, Chik W, Zecchin R, et al. Single-ring posterior left atrial (box) isolation results in a different mode of recurrence compared with wide antral pulmonary vein isolation on long-term follow-up: Longer atrial fibrillation-free survival time but similar survival time free of any atrial arrhythmia. Circ Arrhythm Electrophysiol 2012; 5: 968–977.
• 14.   Huo Y, Gaspar T, Pohl M, Sitzy J, Richter U, Neudeck S, et al. Prevalence and predictors of low voltage zones in the left atrium in patients with atrial fibrillation. Europace 2018; 20: 956–962.
• 15.   Pan NH, Tsao HM, Chang NC, Chen YJ, Chen SA. Aging dilates atrium and pulmonary veins: Implications for the genesis of atrial fibrillation. Chest 2008; 133: 190–196.
• 16.   Friberg L, Tabrizi F, Englund A. Catheter ablation for atrial fibrillation is associated with lower incidence of stroke and death: Data from Swedish health registries. Eur Heart J 2016; 37: 2478–2487.
• 17.   Hunter RJ, McCready J, Diab I, Page SP, Finlay M, Richmond L, et al. Maintenance of sinus rhythm with an ablation strategy in patients with atrial fibrillation is associated with a lower risk of stroke and death. Heart 2012; 98: 48–53.
• 18.   Bunch TJ, Crandall BG, Weiss JP, May HT, Bair TL, Osborn JS, et al. Patients treated with catheter ablation for atrial fibrillation have long-term rates of death, stroke, and dementia similar to patients without atrial fibrillation. J Cardiovasc Electrophysiol 2011; 22: 839–845.
• 19.   Yagishita A, Takahashi Y, Takahashi A, Fujii A, Kusa S, Fujino T, et al. Incidence of late thromboembolic events after catheter ablation of atrial fibrillation. Circ J 2011; 75: 2343–2349.
• 20.   Kanda T, Masuda M, Fujita M, Iida O, Okamoto S, Ishihara T, et al. Comparison of the origin and coupling interval between ectopy with and without atrial fibrillation initiation. J Cardiol 2018; 71: 59–64.
• 21.   Brooks AG, Rangnekar G, Ganesan AN, Salna I, Middeldorp ME, Kuklik P, et al. Characteristics of ectopic triggers associated with paroxysmal and persistent atrial fibrillation: Evidence for a changing role. Heart Rhythm 2012; 9: 1367–1374.