Article ID: CJ-18-0650
Background: Pulmonary vein isolation (PVI) by means of cryoballoon is increasingly being used for the treatment of atrial fibrillation (AF). This study assessed whether the left atrial volume index (LAVI) predicts AF recurrence following PVI by means of 2nd-generation cryoballoon (Cryoballoon Advance; CB-A) when comparing persistent AF (PeAF) and paroxysmal AF (PAF).
Methods and Results: Patients with drug-resistant AF and undergoing preprocedural computed tomography (CT) and PVI with CB-A were included. LAV was estimated from 3D CT images. A total of 417 patients with AF were included (95 PeAF, 322 PAF patients). After a mean of 22.1±9.4 months follow-up, 45/95 (47%) PeAF patients and 254/322 (79%) PAF patients had no recurrence. LAVI was an independent predictor for AF recurrence in PeAF patients (hazard ratio 1.042 per 1 mL/m2; 95% confidence interval 1.006–1.080, P=0.02), but not in PAF patients. In PeAF patients with LAVI ≤61 mL/m2, the freedom from recurrence was 78.5% vs. 22.2% in those with LAVI >61 mL/m2 (hazard ratio 5.771, 95% confidence interval 2.434–13.682, P<0.001), and the mid-term success rate was comparable with PAF patients.
Conclusions: LAVI predicted AF recurrence after PVI using CB-A in PeAF patients but not in PAF patients. If LAVI was ≤61 mL/m2, the mid-term efficacy among PeAF patients was equivalent to that for PAF patients.
The 2nd-generation cryoballoon (CB-A, Arctic Front AdvanceTM, Medtronic, MN, USA) is increasingly being used for the treatment of atrial fibrillation (AF) because of its proven efficacy.1 Recent studies indicate that CB-A ablation offers freedom from arrhythmic recurrence in roughly 80% of patients with paroxysmal AF (PAF) and in 60% of those with persistent AF (PeAF). Previous studies also reported that the left atrial volume index (LAVI) assessed by multidetector computed tomography (CT) is a predictor of AF recurrence after pulmonary vein isolation (PVI).2,3 However, studies focusing on LAVI as a predictor of arrhythmic recurrence following CB-A for persistent AF are sparse. Therefore, we assessed the role of preprocedural LAVI in determining arrhythmic recurrence in a large cohort of patients that underwent CB-A as an index procedure for PVI.
All patients who underwent PVI as the index procedure between June 2012 and March 2015 for symptomatic, drug-refractory PeAF and PAF, defined according to the current guidelines, were consecutively included in this single-center, prospective observational study and clinically followed.4 Long-standing PeAF (LSAF) was defined as PeAF with a duration >12 months before ablation. Exclusion criteria were presence of an intracavitary thrombus, uncontrolled heart failure (HF), previous PVI procedure, lack of preprocedural enhanced cardiac CT imaging and follow-up <12 months. The study design was approved by the institutional ethics committee according to the principles of the Declaration of Helsinki.
Preprocedural ManagementAll patients provided written informed consent prior to the procedure. To exclude the presence of thrombi in the LA appendage (LAA), patients underwent transesophageal echocardiography the day before the procedure, as well as transthoracic examination of the LA, left ventricle, and valvular function. All antiarrhythmic drugs (AAD) were discontinued at least 3 days prior to ablation. For patients under novel anticoagulant agents, our practice is to stop anticoagulation as follows: (1) the last dose of rivaroxaban is given in the morning 1 day prior to the procedure, and (2) the last dose of dabigatran or apixaban is given in the evening 1 day prior to procedure. Warfarin or acenocumarol are not interrupted.
CT Image AnalysisQuantitative assessment of the LA was rendered by the 3D threshold-based method. The anatomy of the LA and PVs and the LAV were identified for all cases. The 2D LA area was semi-automatically traced on each CT slice from the LA roof to the level of the mitral annulus. The PVs were cut at the PV ostia and the LAA was excluded at its base.5 The LAVI was calculated by dividing the LAV by body surface area using the DuBois formula. CT images were analyzed by 2 observers in consensus. Any inconsistencies related to anatomic variants and measurements between the observers were resolved by a third party.
Ablation ProcedureCryoballoon ablation procedure was carried out as previously described in detail.1 Briefly, after gaining LA access, through a steerable 15Fr sheath (FlexCath Advance®, Medtronic), an inner lumen-mapping catheter (Achieve®, Medtronic) was advanced into each PV ostium. Next, a 28-mm CB-A (Arctic Front Advance) was inflated and positioned sequentially at the ostium of each PV. Optimal vessel occlusion was considered as achieved upon selective contrast injection. Cryothermal applications lasted 3 min.6 After the application, an additional freeze was delivered in case of failure to isolate the PV, if the nadir temperature was greater than −35 degrees Celsius and in the occurrence of early LA-PV reconduction. PVI was defined as no PV activity after the applications. In order to avoid right-sided phrenic nerve palsy (PNP), diaphragmatic stimulation was performed by pacing the ipsilateral nerve as previously described.7 Except for the cavotricuspid isthmus line, which was performed if common right atrial flutter was clinically involved together with AF, no additional focal (non-PV trigger) or linear ablation (roof/mitral) was performed during the index procedure. If after PVI, AF did not convert to sinus rhythm, external electrical cardioversion was delivered. During the whole procedure activated clotting time was maintained >250 s.
Postprocedural ManagementOral anticoagulation was started the same evening of the ablation and continued for at least 3 months. Antiarrhythmic therapy was administered for 3 months following the procedure and discontinued if the patient was free of AF relapse.
Follow-upAfter the procedure, all patients underwent clinical evaluation, ECG and a 24-h Holter recording at 1, 3, 6, 12 months. Patients with implanted pacemakers or defibrillators were eligible to be followed by device interrogation. Recurrence of AF was defined as a symptomatic or documented episode >30 s. A blanking period (BP) of 3 months was considered for the study. Repeat ablations were performed in those who had documented symptomatic atrial arrhythmia recurrence beyond the BP. All repeat procedures were performed with an RF irrigated-tip contact-force catheter guided by an electroanatomic-mapping system.
Statistical AnalysisCategorical variables are expressed as absolute and relative frequencies. Continuous variables are expressed as mean±SD or median and interquartile range [IQR] as appropriate. Comparisons of continuous variables were performed with a Student’s t-test or the Mann-Whitney U-test as appropriate. The Chi-square test or Fisher test was used to compare categorical variables. Linear regression analysis was performed to relate LAVI and AF duration from first diagnosis to the index procedure. Factors predicting AF recurrence were identified by univariate and multivariable analyses using a Cox proportional hazard regression model. A receiver-operating characteristic (ROC) curve was constructed to evaluate the performance of LAVI in predicting AF recurrence, and the area under the curve (AUC) was calculated to provide a summary measure of the accuracy of the prediction model. The optimized cutoff point was acquired using the maximal Youden Index. Event-free survival was estimated by the method of Kaplan-Meier and compared by the log-rank test. A two-tailed probability value of 0.05 was deemed significant. All analyses were performed with IBM SPSS Statistics, Version 24.0. (IBM Corp., Armonk, NY, USA).
A total of 417 patients (273 male, 65%; mean age 57.8±13.4 years) were included in the analysis: 95 patients with PeAF and 322 patients with PAF were evaluated. Among the PeAF patients, 11 LSAF patients were involved. The LAVI of the whole patient cohort was 54.2±17.3 mL/m2. At the beginning of the procedure, 85/417 (20%) patients presented with AF. In 6 patients who were initially in sinus rhythm, AF was initiated spontaneously during the procedure. Conversion to sinus rhythm occurred during the procedure in 42/91 (46%) patients. PVI was achieved in all PVs (1,660, 100%). Mean procedural time (from venous access to removal of catheters from the LA) and mean fluoroscopy time were 67.3±19.7 min and 15.3±9.0 min, respectively.
Among the PeAF patients, AF was more frequently present at the beginning of the procedure as compared with PAF patients (69/95 [73%] vs. 15/322 [5%], P<0.001). Conversion during the procedure was achieved in 25/69 (36%) PeAF patients and 17/21 (81%) PAF patients (P=0.001). Additional cavocuspid isthmus ablation was performed in 3/95 (3%) PeAF patients and 5/322 (2%) PAF patients (P=0.39). Table 1 presents the baseline and procedural characteristics of patients.
| PeAF (n=95) |
PAF (n=322) |
P value | |
|---|---|---|---|
| Age (years) | 64.3±10.2 | 55.9±13.7 | <0.001 |
| Male sex | 69 (73) | 204 (63) | 0.11 |
| BMI (kg/m2) | 28.7±5.1 | 26.2±4.3 | <0.001 |
| AF duration (months)† | 48 [12–96] | 16 [6–39] | <0.001 |
| Failed class I or III AADs | 1.64±0.70 | 1.55±0.69 | 0.24 |
| Hypertension | 58 (61) | 124 (39) | <0.001 |
| Diabetes mellitus | 14 (15) | 22 (7) | 0.02 |
| Dyslipidemia | 36 (38) | 109 (34) | 0.47 |
| Heart failure | 14 (15) | 13 (4) | 0.001 |
| CAD | 15 (16) | 20 (6) | 0.006 |
| LVEF (%) | 55.3±9.4 | 58.6±5.9 | 0.001 |
| Follow-up (months) | 20.6±8.6 | 22.6±9.6 | 0.08 |
| LAV (mL) | 139.8±37.2 | 95.7±29.3 | <0.001 |
| LAVI (mL/m2) | 69.1±23.3 | 49.2±14.1 | <0.001 |
| Left common PV | 37 (39) | 86 (29) | 0.08 |
| Right common PV | 2 (2) | 4 (1) | 0.64 |
| Left middle PV | 0 (0) | 1 (0.3) | 1.00 |
| Right middle PV | 18 (19) | 53 (18) | 0.88 |
| AF at procedure | 69 (73) | 15 (5) | <0.001 |
| Procedure time (min) | 69.1±23.3 | 66.8±18.5 | 0.33 |
| Fluoroscopy time (min) | 16.0±11.1 | 15.1±8.2 | 0.40 |
Values are mean±SD, median [interquartile range] or number (%). †From first diagnosis to the index procedure. AADs, antiarrhythmic drugs; AF, atrial fibrillation; BMI, body mass index; CAD, coronary artery disease; CT, computed tomography; LA, left atrium; LAV, LA volume; LAVI, LAV index; LVEF, left ventricular ejection fraction; PAF, paroxysmal AF; PeAF, persistent AF; PV, pulmonary vein.
LAVI was significantly greater in the PeAF compared with the PAF patients (69.1±23.3 mL/m2 vs. 49.2±14.1 mL/m2, P<0.001). Because the LAVI was considered to be involved in AF duration from first diagnosis to the index procedure, we analyzed the correlation and found a weak linear correlation between LAVI and AF persistence (Spearman test: P<0.001, R=0.204; [LAVI (mL/m2)]=[AF duration (months)]*0.06+53.3). In the PeAF patients, 11 LSAF patients showed greater LAVI compared with 84 early PeAF patients (67.1±16.4 mL/m2 vs. 88.9±10.9 mL/m2, P<0.001).
ComplicationsPNP was the most frequent complication, occurring in 24 (6%) patients. In 92% (22/24) of them, PNP resolved completely during the procedure. In the remaining 2 cases, the palsy persisted after discharge but was not present 1 month later. A total of 8 (2%) patients experienced a pseudoaneurysm that required surgical treatment and a TIA occurred in 1 patient on the day after ablation, with no permanent neurological sequelae. Neither pericardial tamponade nor clinically significant pericardial effusion occurred in our study group.
Follow-upA total of 48 (12%) patients had early recurrence (ER) within the first 3 months after the procedure. After a mean follow-up duration of 22.1±9.4 months, 299/417 (72%) patients were free from late recurrence (after the index procedure only). When including redo procedures, the number of patients with freedom from recurrence increased to 329/417 (79%).
The ER rate was 30% (28/95 patients) in the PeAF group and 6% (20/322) in the PAF group (P<0.001). At the end of the follow-up, 47% (45/95) of the patients with PeAF and 79% (254/322) in PAF group were free from late recurrence (P<0.001), considering only the index procedure. Taking redo procedures into account, the freedom from recurrence rate increased to 56 (59%) and 273 (85%) after 44 and 56 redo procedures for PeAF and PAF patients, respectively (P<0.001).
Predictors of Late RecurrenceTable 2 shows the univariate Cox proportional hazard modeling of AF recurrence. Age, body mass index, duration of AF from first diagnosis to the index procedure, PeAF, LSAF, ER, hypertension, diabetes mellitus, HF, LAV and LAVI were univariate predictors of AF recurrence. In the multivariable analysis, only LAVI was independently associated with outcomes (hazard ratio [HR] per 1 mL/m2 increase 1.063; 95% confidence interval [CI] 1.017–1.110, P=0.007).
| Variable | HR | 95% CI | P value |
|---|---|---|---|
| Univariate analysis | |||
| Age | 1.016 | 1.001–1.030 | 0.04 |
| Male sex | 0.936 | 0.643–1.361 | 0.73 |
| BMI | 1.047 | 1.008–1.087 | 0.02 |
| AF duration | 1.003 | 1.000–1.006 | 0.002 |
| PeAF | 3.261 | 2.262–4.703 | <0.001 |
| LS PeAF | 6.335 | 3.287–12.209 | <0.001 |
| Conversion to SR† | 0.623 | 0.326–1.191 | 0.15 |
| Early recurrence | 3.284 | 2.159–4.995 | <0.001 |
| Hypertension | 1.438 | 1.002–2.063 | 0.049 |
| Diabetes mellitus | 1.795 | 1.060–3.041 | 0.03 |
| Dyslipidemia | 1.061 | 0.730–1.541 | 0.76 |
| Heart failure | 1.993 | 1.120–3.547 | 0.02 |
| CAD | 1.416 | 0.795–2.519 | 0.24 |
| LAV | 1.016 | 1.011–1.021 | <0.001 |
| LAVI | 1.034 | 1.024–1.044 | <0.001 |
| Left common PV | 0.901 | 0.602–1.348 | 0.61 |
| Right middle PV | 0.969 | 0.598–1.572 | 0.90 |
| Multivariable analysis | |||
| Age | 0.991 | 0.968–1.015 | 0.46 |
| BMI | 1.046 | 0.988–1.108 | 0.12 |
| AF duration | 1.001 | 0.997–1.004 | 0.59 |
| PeAF | 1.538 | 0.779–3.036 | 0.22 |
| LS PeAF | 1.557 | 0.684–3.547 | 0.29 |
| Early recurrence | 1.382 | 0.774–2.467 | 0.27 |
| Hypertension | 0.809 | 0.483–1.355 | 0.42 |
| Diabetes mellitus | 1.063 | 0.529–2.134 | 0.86 |
| Heart failure | 1.622 | 0.831–3.164 | 0.16 |
| LAV | 0.985 | 0.966–1.005 | 0.15 |
| LAVI | 1.063 | 1.017–1.110 | 0.007 |
†Conversion from AF to SR during the procedure. CI, confidence interval; HR, hazard ratio; LS PeAF, long-standing PeAF; SR, sinus rhythm. Other abbreviations as in Table 1.
In the PeAF patients, univariate analysis showed that HF, LAV and LAVI were associated with late recurrence, and only LAVI maintained statistical significance in the multivariable model (HR per 1 mL/m2 increase 1.042; 95% CI 1.006–1.080, P=0.02) (Table 3). However, in PAF patients, ER (HR 3.577; 95% CI 1.875–6.824, P<0.001) and hypertension (HR 1.611; 95% CI 1.002–2.592, P=0.049) were identified in the univariate analysis, albeit the multivariable analysis showed that only ER was a predictor for late AF recurrence (HR 3.462; 95% CI 1,813–6.610, P<0.001) (Table 3). LAVI of PAF patients was not associated with late recurrence even in the univariate analysis.
| Variables | PeAF | PAF | ||||
|---|---|---|---|---|---|---|
| HR | 95% CI | P value | HR | 95% CI | P value | |
| Univariate analysis | ||||||
| Age | 1.017 | 0.998–1.047 | 0.26 | 0.999 | 0.981–1.016 | 0.88 |
| Male sex | 0.734 | 0.405–1.332 | 0.31 | 0.895 | 0.551–1.455 | 0.65 |
| BMI | 1.032 | 0.975–1.092 | 0.28 | 1.014 | 0.961–1.070 | 0.60 |
| AF duration | 1.003 | 1.000–1.007 | 0.06 | 1.002 | 0.997–1.006 | 0.53 |
| Conversion to SR† | 1.061 | 0.545–2.068 | 0.86 | 0.217 | 0.014–3.476 | 0.28 |
| Early recurrence | 1.589 | 0.891–2.832 | 0.116 | 3.577 | 1.875–6.824 | <0.001 |
| Hypertension | 0.658 | 0.376–1.152 | 0.14 | 1.611 | 1.002–2.592 | 0.049 |
| Diabetes mellitus | 1.547 | 0.751–3.189 | 0.24 | 1.495 | 0.684–3.268 | 0.31 |
| Dyslipidemia | 1.043 | 0.589–1.847 | 0.89 | 1.035 | 0.632–1.696 | 0.89 |
| Heart failure | 2.093 | 1.068–4.099 | 0.03 | 0.627 | 0.154–2.561 | 0.52 |
| CAD | 1.299 | 0.648–2.603 | 0.46 | 0.688 | 0.216–2.190 | 0.53 |
| LAV | 1.016 | 1.008–1.024 | <0.001 | 1.005 | 0.997–1.014 | 0.21 |
| LAVI | 1.038 | 1.022–1.055 | <0.001 | 1.009 | 0.992–1.026 | 0.32 |
| Left common PV | 0.953 | 0.538–1.689 | 0.87 | 0.702 | 0.392–1.256 | 0.23 |
| Right middle PV | 1.620 | 0.845–3.108 | 0.15 | 0.631 | 0.300–1.327 | 0.23 |
| Multivariable analysis | ||||||
| Early recurrence | 3.462 | 1.813–6.610 | <0.001 | |||
| Hypertension | 1.560 | 0.969–2.511 | 0.07 | |||
| Heart failure | 1.770 | 0.898–3.490 | 0.10 | |||
| LAV | 0.997 | 0.980–1.015 | 0.76 | |||
| LAVI | 1.042 | 1.006–1.080 | 0.02 | |||
†Conversion from AF to SR during the procedure. Abbreviations as in Tables 1,2.
The predictive performance of LAVI was calculated by ROC analysis. In the whole cohort of patients, ROC analysis showed an AUC of 0.674. However, the ROC curve in PeAF patients demonstrated a performance of 0.775 (Figure 1); the optimal calculated cutoff value for predicting recurrences was 61 mL/m2 of LAVI. In PAF the ROC curve had an AUC of 0.534.
Receiver-operating characteristic curve predicting AF recurrence from left atrial volume index of PeAF patients. Area under the curve value is included. PeAF, persistent atrial fibrillation.
Freedom from AF was significantly different among the patients with PAF, PeAF with LAVI ≤61 mL/m2 and PeAF with LAVI >61 mL/m2 (Figure 2). In PeAF, 78.5% of patients with LAVI ≤61 mL/m2 were free of recurrence as compared with 22.2% of those with LAVI >61 mL/m2 (log-rank test P<0.001, unadjusted HR 6.060, CI 2.567–14.307, P<0.001 and adjusted HR 5.771, CI 2.434–13.682, P<0.001). The characteristics of the 2 groups showed no significant differences except for duration from first diagnosis of AF to the index procedure, which was longer in PeAF patients with LAVI >61 mL/m2 (Table 4). Both PAF and PeAF patients with LAVI ≤61 mL/m2 had similar outcomes (P=0.69, Figure 2).
Kaplan-Meier curves showing survival free of atrial fibrillation recurrence during follow-up among PeAF patients with LAVI ≤61 mL/m2, PeAF patients with LAVI >61 mL/m2 and PAF patients after a single procedure and (off antiarrhythmic drugs). AF, atrial fibrillation; LAVI, left atrial volume index; PAF, paroxysmal atrial fibrillation; PeAF, persistent atrial fibrillation.
| LAVI | P value | ||
|---|---|---|---|
| ≤61 mL/m2 | >61 mL/m2 | ||
| n | 32 | 63 | |
| Age (years) | 61.4±11.3 | 65.7±9.3 | 0.051 |
| Male sex | 25 (78) | 44 (70) | 0.47 |
| BMI (kg/m2) | 27.8±5.1 | 29.2±5.1 | 0.21 |
| AF duration (months) | 19.5 [6.25–65.5] | 60 [19–120] | 0.005 |
| LS PeAF | 0 (0) | 11 (18) | 0.01 |
| Failed class I or III AADs | 1.69±0.69 | 1.62±0.71 | 0.83 |
| Hypertension | 19 (59) | 39 (62) | 0.83 |
| Diabetes mellitus | 2 (6) | 12 (19) | 0.13 |
| Dyslipidemia | 11 (34) | 25 (40) | 0.66 |
| Heart failure | 2 (6) | 12 (19) | 0.13 |
| CAD | 3 (9) | 12 (19) | 0.37 |
| LVEF (%) | 57.6±9.3 | 54.1±9.3 | 0.09 |
| Follow-up (months) | 20.5±8.2 | 20.7±8.9 | 0.94 |
| LAV | 103.0±17.9 | 158.5±29.7 | <0.001 |
| LAVI (mL/m2) | 50.9±7.3 | 79.2±12.3 | <0.001 |
| Left common PV | 14 (44) | 23 (37) | 0.51 |
| Right common PV | 0 (0) | 2 (3) | 0.55 |
| Right middle PV | 3 (9) | 15 (24) | 0.11 |
Abbreviations as in Tables 1,2.
The main findings of the present study were: (1) LAVI was a significant preprocedural predictor of late AF relapse in patients with PeAF undergoing PVI with the 2nd-generation cryoballoon and, importantly, (2) the success rate after the index CB-A was similar between PeAF patients with LAVI ≤61 mL/m2 and PAF patients.
Predictors of Recurrence in Patients With PeAFLAVI significantly affected late success in patients with PeAF treated with the 2nd-generation cryoballoon. To the best of our knowledge, our study is the first to prove this. Previously, Abecasis et al reported that LAV >145 mL had significantly higher recurrence rates of arrhythmia in 99 patients with AF at a mean follow-up period of 17 months after PVI with radiofrequency ablation.2 Interestingly, in our study, we demonstrated that LAVI was a predictor of recurrence only in patients with PeAF; in PAF patients its utility is limited.
Atrial structural and electrical remodeling is directly related to the presence and maintenance of atrial arrhythmias.8,9 The magnitude of atrial fibrosis is proportional to the duration of AF (AF begets AF) and is the substrate for ectopic electrical activity and reentry mechanisms.10,11 The aforementioned steps are intrinsic to PeAF and might contribute to arrhythmic recurrence after ablation procedures. In contrast, in PAF, the structural and electrical alterations can be restored if the exposure to AF is <7 days.12,13 This might explain why LAVI failed to predict AF recurrence in PAF.
We did not identify ER as a predictor of late recurrence in PeAF patients, although previous studies reported ER as a strong predictor of recurrence after PVI using a cryoballoon.14,15 Reversible arrhythmia mechanisms related to the ablated lesion may cause ER,16 because of the acute inflammatory response to cryo-energy as well as modifications of the cardiac autonomic nervous system.15 Probably owing to the small number of the PeAF patients, this study did not have enough power to show a significant difference in ER among the PeAF patients.
PVI Using CB-A in PeAF Patients With Small LAVIThe mid-term clinical outcomes after PVI using CB-A in PeAF patients with LAVI ≤63 mL/m2 were similar to those in PAF patients. This finding has an intriguing clinical implication; namely, that even in PeAF patients, if their preprocedural CT images show a smaller LAVI, CB-A has a high success rate with an AF-free survival similar to that for PAF patients.
LA remodeling is time-dependent. Small-scale remodeling limits interstitial fibrosis in the LA and enables reverse remodeling following removal of the stressors.17 It may derive a postulate that there is a threshold for deciding controllability of atrial tachyarrhythmias after catheter ablation for PeAF. The CONFIRM trial reported that 97% of sustained AF showed localized rotors and focal impulse in the atria.18 Focal impulse and rotor modification (FIRM) alone achieved AF termination in 56% patients with AF, and FIRM-guided ablation significantly increased long-term efficacy free from AF recurrence. In other studies, ganglion plexus (GP) ablation has been advocated for control of AF. GP ablation with additional PVI decreased the inducibility of sustained AF from 68% to 17%.19 The 28-mm CB-A makes wide cryothermal lesions around the PV ostia, which might partially cover the area of the rotors, focal impulses and the GPs. This hypothesis can explain our study result that the mid-term success rate after cryoballoon ablation in PeAF patients with LAVI ≤61 mL/m2 was similar to that for PAF patients. In contrast, a dilated LA might harbour non-PV sources away from the PV ostia, which would be less likely covered by the effect of the 28-mm CB-A. In addition to the greater amount of AF substrate in patients with a larger LA, associated enlarged size of the PVs may explain the arrhythmogenesis after PVI with CB-A. Compared with standard RF ablation, the fixed size of the CB might create its cryothermal lesions too distally in the case of dilated PVs.
Study LimitationsThis study was a single-center trial conducted in a relatively small number of patients with PeAF. Further larger studies are necessary to confirm our findings. The AF recurrence rate was evaluated with Holter ECG and in only 10% of patients with implanted devices, therefore asymptomatic episodes might be underestimated. Furthermore, as data regarding the isolation area of PVs after CB-A were missing, our study did not provide more insight to the reason why a larger LAVI presents a worse outcome.
Cardiac CT-based LAVI was associated with late AF recurrence. Of note, if the LAVI was ≤61 mL/m2, mid-term AF-free survival after CB ablation for PeAF patients was noninferior compared with PAF patients.
We acknowledge all of the investigators in the present study, including E.S. for data collection and study support, J.S. for protocol development, Juan-Pabro Abugattas for statistical support, and G.-B.C. for manuscript development.
C.deA. received compensation for teaching purposes and proctoring from AF solutions, Medtronic, and is a member of the steering committee ETNA-AF Europe Daiichi Sankyo Europe. P.B. received speaker’s fees from Biotronik, Medtronic. G.-B.C. receives compensation for teaching purposes and proctoring from AF solutions, Medtronic. C.deA. received research grants on behalf of the Centre from Biotronik, Medtronic, St. Jude Medical Abbot, Livanova, Boston Scientific. K.T. previously received an educational grant from the Japanese Heart Rhythm Society. J.-P.A. is currently receiving an educational grant from St. Jude Medical for the Postgraduate in Cardiac electrophysiology and Pacing academic course.
