Author + information
- Received October 31, 2016
- Revision received January 17, 2017
- Accepted January 19, 2017
- Published online June 19, 2017.
- Moloy Das, MBBSa,b,
- Gareth J. Wynn, MD(Res)a,c,
- Yawer Saeed, MBBSa,
- Sean Gomes, MBBSa,d,
- Maureen Morgan, BSca,
- Christina Ronayne, BSca,
- Laura J. Bonnett, PhDe,
- Johan E.P. Waktare, MDa,
- Derick M. Todd, MDa,
- Mark C.S. Hall, MDa,
- Richard L. Snowdon, MDa,
- Simon Modi, MBBSa and
- Dhiraj Gupta, MDa,∗ ()
- aInstitute of Cardiovascular Medicine and Science, Department of Cardiology, Liverpool Heart and Chest Hospital, Liverpool, United Kingdom
- bDepartment of Cardiology, Freeman Hospital, Newcastle upon Tyne, United Kingdom
- cThe Royal Melbourne Hospital, Melbourne, Australia
- dThe Prince of Wales Hospital, Sydney, Australia
- eDepartment of Biostatistics, University of Liverpool, Liverpool, United Kingdom
- ↵∗Address for correspondence:
Dr. Dhiraj Gupta, Department of Cardiology, Liverpool Heart and Chest Hospital, Thomas Drive, Liverpool L14 3PE, United Kingdom.
Objectives The goal of this study was to determine whether a strategy of early re-isolation of pulmonary vein (PV) reconnection in all patients, regardless of symptoms, would reduce the recurrence of atrial fibrillation (AF) and improve quality of life.
Background Lasting pulmonary vein isolation (PVI) remains elusive. PV reconnection is strongly linked to the recurrence of arrhythmia.
Methods A total of 80 patients with paroxysmal AF were randomized 1:1 after contact force-guided PVI to receive either standard care or undergo a repeat electrophysiology study after 2 months regardless of symptoms (repeat study). At the initial procedure, PVI was demonstrated by entrance/exit block and adenosine administration after a minimum 20-min wait. At the repeat study, all sites of PV reconnection were re-ablated. Patients recorded electrocardiograms daily and whenever symptomatic for 12 months using a handheld monitor. Recurrence was defined as ≥30 s of atrial tachyarrhythmia (AT) after a 3-month blanking period. The Atrial Fibrillation Effect on Quality-of-Life Questionnaire was completed at baseline and at 6 and 12 months.
Results All 40 patients randomized to repeat study attended for this after 62 ± 6 days, of whom 25 (62.5%) had reconnection of 41 (26%) PVs. There were no complications related to these procedures. Subjects recorded a total of 32,203 electrocardiograms (380 [335 to 447] per patient) during 12.6 (12.2 to 13.2) months of follow-up. AT recurrence was significantly lower for the repeat study group (17.5% vs. 42.5%; p = 0.03), as was AT burden (p = 0.03). Scores on the Atrial Fibrillation Effect on Quality-of-Life Questionnaire were higher in the repeat study group at 6 months (p < 0.001) and 12 months (p = 0.02).
Conclusions A strategy of routine repeat assessment with re-isolation of PV reconnection improved freedom from AT recurrence, AT burden, and quality of life compared with current standard care. (The Effect of Early Repeat Atrial Fibrillation [AF] on AF Recurrence [PRESSURE]; NCT01942408)
Since the pulmonary veins (PVs) were first identified as the primary sources of initiating triggers of paroxysmal atrial fibrillation (AF) (1), catheter ablation to achieve pulmonary vein isolation (PVI) has become the cornerstone of AF ablation (2). Even in more persistent forms of AF, additional atrial ablation does not improve outcomes compared with PVI alone (3,4). However, creating durable PVI is difficult to achieve, with contemporary studies demonstrating that up to two-thirds of patients after a successful PVI procedure have PV reconnection at a protocol-mandated repeat electrophysiology study (5,6). PV reconnection is well-recognized as the leading cause of atrial tachyarrhythmia (AT) recurrence after PVI (7–10), which occurs in approximately 30% to 40% of patients after a single procedure (11–13). AT recurrences such as these significantly affect quality of life (14,15).
Currently, patients re-presenting with AT recurrence commonly undergo repeat ablation, and this approach has been shown to increase success rates (8). However, there can be a significant delay between symptomatic recurrence and the second procedure, during which time quality of life can be affected. While clinical success rates have remained relatively static, the safety of AF ablation has improved significantly over recent years (16,17). Given the high rates of late PV reconnection still seen, we designed the PRESSURE (Pulmonary vein RE-Isolation as a routine Strategy: a SUccess Rate Evaluation) study to test the hypothesis that a strategy of assessment and re-isolation of all PV reconnection 2 months after the initial AF ablation, irrespective of symptoms, would reduce the recurrence of AT and improve quality of life.
Patients and Methods
Consecutive patients were recruited from a single tertiary center in the United Kingdom. Study inclusion criteria were as follows: a current pattern of paroxysmal AF (defined as electrocardiogram [ECG]-proven episodes of AF that are self-limiting and last <7 days on each occasion or which were cardioverted electrically or pharmacologically <48 h from onset), age >18 years, and scheduled for catheter ablation using radiofrequency energy. Exclusion criteria were: previous ablation procedure for AF, prosthetic mitral valve replacement, severe structural cardiac abnormality, infiltrative cardiomyopathy, severe left ventricular systolic dysfunction (ejection fraction <35%), and pregnancy. All patients provided written informed consent, and the study was approved by the UK National Research Ethics Service and the Institutional Research Committee.
Patients were enrolled before their initial ablation procedure, but randomization was delayed until immediately after the procedure to avoid possible bias or influence on the ablation. Randomization was in a 1:1 ratio using a custom-written 2- and 4-block randomization program to either standard care (SC) or to undergo a repeat electrophysiology study after 2 months, irrespective of symptoms (repeat study group). Due to the nature of the study design, neither the patients nor the clinicians were blinded to randomization, but all study analyses were performed by blinded observers. A diagram defining study time periods is shown in Figure 1.
Initial PVI procedure
Patients taking amiodarone stopped this drug a minimum of 2 months before their PVI procedure; if this stoppage was not possible, the patient was excluded. All other antiarrhythmic drugs were stopped 5 days beforehand. Patients taking warfarin continued this drug peri-procedurally, with an international normalized ratio of 2.0 to 3.5 considered acceptable. Non–vitamin K anticoagulant agents were stopped 24 to 48 h before the procedure.
PVI was performed under conscious sedation with fentanyl and midazolam or general anesthesia in a standard fashion. Vascular access was gained under direct ultrasound guidance, as is standard in our institution (18). A 3-dimensional navigation system (CARTO 3, Biosense Webster, Inc., Diamond Bar, California) was used in all cases to create an electroanatomic map of the left atrium and, where possible, this map was integrated with a computed tomography or magnetic resonance imaging reconstruction of the left atrium (CartoMerge, Biosense Webster, Inc.). PVI was performed in a wide-area circumferential ablation pattern by using a 3.5-mm irrigated tip contact force-sensing radiofrequency ablation catheter (Thermocool SmartTouch, Biosense Webster, Inc.). Power settings were 25 to 30 W for the posterior wall and 30 to 35 W for other regions. A contact force of 5 to 40 g with application duration of 20 to 40 s was targeted at each site, aiming for local signal attenuation >80%. Automated lesion tagging (VisiTag, Biosense Webster, Inc.) was used to mark the location of each lesion, using standardized settings as follows: catheter position stability—minimum time 10 s, maximum range 2 mm; force over time—time 30%, minimum force 5 g; and lesion tag size—2 mm. Operators aimed for contiguous lesions around the wide-area circumferential ablation circle. Cavotricuspid isthmus ablation was permitted in patients with documented typical right atrial flutter, but no additional left atrial ablation was performed, and no attempt was made to look for extra-PV triggers. AF that persisted after PVI was terminated by using electrical cardioversion.
Acute PVI was confirmed by demonstrating entry and exit block with a 20-pole circular mapping catheter (Lasso NAV Eco, Biosense Webster, Inc.) placed sequentially in each of the PVs. Ongoing PVI was confirmed a minimum of 20 min after isolation of that ipsilateral PV pair, with intravenous adenosine boluses (12 to 18 mg) administered to unmask sites of dormant conduction. Further ablation was performed at sites of overt or unmasked reconnection to re-isolate the PVs.
Post-procedure antiarrhythmic drug therapy
Pre-procedure antiarrhythmic medications, including beta-blockers and rate-limiting calcium channel-blockers, were restarted after the procedure and were stopped after 4 weeks. Subsequent re-initiation of therapy by the attending physician to treat recurrences of clinical arrhythmia was permitted but these medications were stopped again at the end of the 3-month blanking period.
Repeat study group
A repeat electrophysiology study was performed 2 months after the initial procedure in all patients in the repeat study group. Any re-initiated antiarrhythmic medications were stopped again 5 days beforehand. The repeat study was performed by using 3-dimensional mapping in the same way as for the initial procedure. Each PV was assessed for reconnection by using a 20-pole circular catheter. All identified sites of reconnection were ablated to re-isolate the PVs. As for the initial procedure, no additional left atrial ablation beyond PVI was performed. As pre-specified, a Data Monitoring and Safety Committee review was held after one-half of the repeat procedures in the repeat study group had been performed.
Patients in the SC arm did not undergo a routine repeat electrophysiology study and, as per international guidelines, repeat ablation was not performed within the first 3 months (2). Management was otherwise identical to the repeat study group, with options of antiarrhythmic drugs or repeat ablation available as per standard clinical care for patients with symptomatic AT after the 3-month blanking period.
All patients were provided with a validated portable ECG monitor (Omron HCG-801-E, Omron Healthcare, Kyoto, Japan) (19) and, having been trained in its use, were instructed to self-record a 30-s ECG every day and additionally whenever they experienced symptoms. ECG recordings were downloaded at each follow-up visit and were analyzed for the presence and dates of any AT by experienced clinicians (Y.S. and S.G.) blinded to treatment allocation.
Follow-up visits were arranged after 6 weeks and 3, 6, and 12 months. Any antiarrhythmic medications restarted for symptomatic documented AT during the blanking period were stopped again 3 months after the initial ablation. The validated Atrial Fibrillation Effect on Quality-of-Life Questionnaire (AFEQT), which includes 20 questions across 3 domains (symptoms, daily activities, and treatment concerns) using a 7-point Likert scale, was completed at baseline and at 6 and 12 months (20).
The primary outcome measure was AT recurrence occurring after a 3-month blanking period following the initial ablation procedure. As per current guidelines, AT recurrence was defined as documented AF, atrial flutter, or atrial tachycardia lasting ≥30 s (2,21).
Pre-specified secondary outcome measures were as follows: quality of life 6 and 12 months after initial ablation; time to first AT recurrence; AT burden during the primary outcome period; re-initiation of antiarrhythmic medication; and comparison of major complication rates. Procedure-related complications consisted of cardiac tamponade, stroke/transient ischemic attack, myocardial infarction, severe PV stenosis, phrenic nerve paralysis, esophageal perforation or atrio-esophageal fistula, major vascular complications, and death. Definitions were as per international guidelines (2).
A formal sample size calculation was performed assuming a 12-month single-procedure success rate of 64% in the SC group, based on the ThermoCool AF trial results (22), and 90% in the repeat study group. Although difficult to predict due to the novel trial design, the estimated success rate in the repeat study group was based on the A4 study, in which up to 2 additional procedures were allowed within a 90-day blanking period for patients with early recurrence; this approach resulted in 89% freedom from AF at 12 months after a mean of 1.8 ± 0.8 procedures (23). Using an alpha error of 0.05 and a beta of 20% (80% power), the number of patients required was 76 (38 in each group). Allowing for 5% attrition yielded a total sample size of 80. The study was intentionally underpowered to detect a difference between groups of <25% because it was felt that the magnitude of the additional intervention required a clear and considerable reduction in AT recurrence to be considered acceptable in clinical practice.
All endpoints were examined by means of an intention-to-treat analysis. Categorical variables and endpoints (including the primary endpoint) were summarized as counts (percentages) and compared with the chi-square or Fisher exact test as appropriate. Continuous variables were checked for normality by visual inspection of the relevant histogram. Where normality was an unreasonable assumption, transformations of the data were attempted (e.g., log and exponential). Continuous variables that were normally distributed were expressed as mean ± SD and were compared by using Student t tests. Variables that were not normally distributed despite transformation were expressed as a median (interquartile range) and were compared by using Mann-Whitney U tests. Time to first AT recurrence was assessed via Cox’s proportional hazard model, and a comparison was made between groups by using the log-rank test. All tests were 2-sided, and a p value <0.05 was considered statistically significant.
A CONSORT diagram showing recruitment to the study is shown in Figure 2. Patient demographic characteristics for the 80 subjects randomized in the study are provided in Table 1. The median CHA2DS2-VASc score was higher in the SC group, although this finding would have been influenced in part by the higher proportion of female subjects in this group. There were no other statistically significant differences between groups at baseline.
Initial procedural characteristics
There were no significant differences in procedural characteristics between groups (Table 2). In 2 patients, the right inferior PV could not be safely isolated at the posterior aspect due to esophageal temperature rise, yielding an overall PVI rate of 99.4%. Of the remaining PVs, acute reconnection (spontaneous or adenosine-mediated) was identified in 56 (17.6%) PVs in 38 (47.5%) patients. There was no difference in acute reconnection rates between groups (SC: 32 [20%] PVs in 18 [45%] patients; repeat study: 24 [15%] PVs in 20 [50%] patients; p = 0.30 and p = 0.82, respectively). All identified sites of acute reconnection were successfully ablated.
Repeat study procedural characteristics
All 40 patients randomized to the repeat study group underwent the repeat procedure at a mean interval of 62 ± 6 days after the initial ablation. Thirty-eight patients were in sinus rhythm at the start of the repeat procedure, with 1 in AF and 1 in atrial flutter. The repeat study procedure duration was 80 (61 to 109) min, fluoroscopy time was 8.3 (6.8 to 12.0) min, and the dose was 918 (498 to 1,756) cGy/m2. The 2 patients in whom the right inferior PV could not be isolated at the initial procedure were both subsequently randomized to the repeat study group. Including these 2 patients, late PV reconnection was identiﬁed in 25 (62.5%) patients, affecting 41 (26%) PVs. The distribution of reconnected PVs was as follows: left superior, 6; left inferior, 14; right superior, 8; and right inferior, 13. All these PVs were successfully re-isolated, including the 2 right inferior PVs not isolated at the initial procedure because the position of the esophagus was found to have changed at the repeat study. The median ablation time for re-isolation in these 25 patients was 5.1 (3.6 to 9.6) min.
In the SC group, 9 (22.5%) patients subsequently underwent a redo procedure for symptomatic arrhythmia recurrence after a median interval of 210 (173 to 233) days from the initial PVI. One of these redo procedures was performed in a patient with frequent symptomatic bursts of atrial ectopy but no sustained arrhythmia ≥30 s recorded on ECG; this patient was therefore categorized as not having experienced the primary endpoint. Eight patients were in sinus rhythm at the start of the procedure, with 1 in atypical atrial flutter. Eight (89%) of these 9 patients had PV reconnection, affecting 18 (50%) PVs. The median procedure and ablation times were 115 (76 to 170) min and 12.3 (6.9 to 15.6) min, respectively, and the fluoroscopy time and dose were 9.9 (6.0 to 14.8) min and 1,024 (254 to 1,709) cGy/m2.
No patient in either group underwent a third left atrial ablation procedure during follow-up.
Patients were followed up clinically for 382 (372 to 402) days (12.6 [12.2 to 13.2] months), with no difference in follow-up duration between groups. ECG follow-up, defined as the number of days from initial ablation to the date of the last self-recorded ECG, was 380 (367 to 400) days and also did not differ between groups (Table 2).
Patients recorded a total of 32,203 ECGs (380 [335 to 447] per patient) during follow-up. Of these, 22,789 ECGs were recorded during the primary outcome period, with no difference between groups (278 [222 to 326] ECGs per patient for the SC group vs. 274 [242 to 315] for the repeat study group; p = 0.81).
During 1 year of follow-up, significantly fewer patients in the repeat study group (n = 7 [17.5%]) experienced the primary endpoint compared with the SC group (n = 17 [42.5%]; p = 0.03). As detailed earlier, 8 of these 17 SC group patients underwent a redo procedure, with the remainder opting to continue with medical management due to either low symptom burden or adequate symptom control with previously ineffective antiarrhythmic medication. One (12.5%) of these 8 patients had further recurrence of AT during the remaining follow-up (5.6 [4.5 to 6.6] months). In the repeat study group, 3 (12%) of the 25 patients with PV reconnection at repeat study had recurrence of AT compared with 4 (27%) of those without.
Time to first recurrence
Figure 3 displays Kaplan-Meier curves for the 2 groups, with the final follow-up date taken as the date of the last self-recorded ECG for censored cases. The log-rank test was statistically significant (p = 0.02). Including a time-dependent covariate for the randomized group showed that the proportional hazards assumption required for the Cox proportional hazards model was valid (p for interaction = 0.78). The Cox model including only the randomized group produced a hazard ratio of 0.35 (95% confidence interval: 0.15 to 0.86; p = 0.02).
The CHA2DS2-VASc score and left atrial diameter were assessed as co-variants of potential clinical interest. Using forward and backward selection methods (via the Akaike information criterion ) led to a model that included only the randomized group (which was forced into the model). Patients in the repeat study group were 64% less likely to have a recurrence than those in the SC group (hazard ratio: 0.36; 95% confidence interval: 0.15 to 0.88; p = 0.02).
Kaplan-Meier curves for the 2 groups (including clinical redo procedures for documented AT in the SC group patients) are shown in Figure 4. The log-rank p value was no longer significant.
AT burden during the primary outcome period
Total group AT burden, defined as the number of patient-days on which AT was documented during the primary outcome period, was markedly lower in the repeat study group than in the SC group (91 vs. 127 days), and the median number of days per patient was significantly lower (0 [0 to 0] vs. 0 [0 to 3]; p = 0.03).
Procedure-related adverse events
No patient died or experienced a stroke during the study. One patient developed cardiac tamponade during the initial procedure, which was successfully drained percutaneously without clinical sequelae. This patient was withdrawn from the study before randomization, but the complication was included in the overall rate. Another patient developed right phrenic nerve palsy after the initial PVI; this condition resolved completely during the follow-up period. The overall serious complication rate associated with 81 initial procedures was 2.5%.
None of the 40 repeat procedures in the repeat study group were associated with any adverse event. One patient in the SC group who underwent repeat ablation for clinical recurrence experienced a post-procedural transient ischemic attack (possibly related to nonabsorption of her oral anticoagulant agent due to vomiting). Considering all procedures, the overall serious complication rates for the 2 groups were not different (repeat study group: 1 in 40 patients [2.5%] in 80 procedures [1.25%]; SC group: 1 in 40 patients [2.5%] in 49 procedures [2.0%]; p > 0.99 for both).
Baseline AFEQT scores did not differ between groups (repeat study group: 46.3 [36.1 to 69.1]; SC group: 47.7 [30.1 to 73.3]; p = 0.90). At 6 months after the initial PVI procedure, AFEQT scores were significantly higher in the repeat study group than in the SC group (88.0 [79.6 to 96.3] vs. 65.7 [49.5 to 82.9]; p < 0.001), and this difference was maintained at 12 months (91.9 [79.5 to 99.1] vs. 77.3 [67.6 to 94.5]; p = 0.02).
AT-related urgent hospital admissions
Excluding elective admissions for repeat AF ablation in both groups, patients in the SC group had markedly more urgent admissions to the hospital relating to AT, either as emergency presentations or for electrical cardioversion (11 vs. 3), with a trend toward significance (p = 0.10).
Re-initiation of medication
Significantly fewer patients in the repeat study group had antiarrhythmic medication (including beta-blockers and calcium-channel blockers) re-initiated in the primary outcome period for symptomatic clinical recurrence (either on a regular or “pill-in-the-pocket” basis) than in the SC group (4 of 40 [10%] vs. 13 of 40 [32.5%]; p = 0.03).
Early recurrence of AT
A planned post hoc analysis of the relationship between early recurrence of AT and the primary endpoint was performed. As shown in Figure 1, the first 4 weeks after ablation were regarded as an absolute blanking period according to previously published data, suggesting this time frame is more reflective of the “true” blanking period (25). The early recurrence period was pre-specified as the period from 4 weeks to the repeat study date (repeat study group) or the end of the blanking period (SC group).
In the SC group, 14 patients had early recurrence, of whom 12 (86%) went on to experience the primary endpoint. This proportion was significantly reduced in the repeat study group (5 [38%] of 13 with early recurrence; p = 0.02). Comparatively, for those without early recurrence (either no recurrence or recurrence confined only to the first 4 weeks), the risk of reaching the primary endpoint was low (7 of 53 [13%]), and the difference between groups did not reach significance (SC group: 5 of 26 [19%]; repeat study group: 2 of 27 [7%]; p = 0.25). The addition of early recurrence to the multivariable model was significant, with patients experiencing early recurrence beyond 4 weeks being nearly 10 times more likely to reach the primary endpoint than those without early recurrence (hazard ratio: 9.75; 95% confidence interval: 3.94 to 24.15; p < 0.001). The addition of early recurrence also increased the significance of the randomized group (p = 0.005).
In this randomized controlled trial, a strategy of re-assessment and ablation of PV reconnection after 2 months, irrespective of symptoms, reduced AT recurrence and burden and improved quality of life compared with current SC, where repeat procedures are only performed for arrhythmia recurrence after an initial blanking period. Although previous data have suggested the potential benefit of early re-intervention in symptomatic patients (26), to our knowledge, the present study is the first to randomize patients regardless of symptoms to PV re-assessment and re-intervention.
Freedom from AT
Our study involved intensive ECG monitoring, which is recognized as resulting in higher AT detection rates (27), and stringent arrhythmia duration criteria of ≥30 s, as per international guidelines for clinical trials (2). AT recurrence occurred in 42.5% of patients in the SC group, which is similar to single-procedure success rates in other contemporary studies (12,13). Comparatively, routine early re-intervention resulted in a success rate of 82.5% with the same degree of intensive monitoring. Previous research has shown that 10% to 25% of patients may have non-PV triggers (1,28,29), and we did not attempt to look for or ablate these. It is also possible that further PV reconnection after re-ablation may have contributed to recurrence in some cases. Nevertheless, early re-intervention conferred a highly significant absolute improvement in freedom from AT of 25%.
After including clinical redo procedures performed for documented AT recurrence in the SC group patients, there was no longer a significant difference in freedom from AT between groups. However, the study was not powered to detect such a difference and, furthermore, follow-up was very short for those patients who underwent a clinical redo procedure, introducing a discrepancy between follow-up periods for the 2 groups. Perhaps more importantly, definitive treatment (the redo procedure) was significantly delayed in these patients compared with the repeat study group (210 vs. 62 days), resulting in more AT burden (with consequently more hospital admissions and cardioversions) and more re-initiation of medication in the intervening period. These factors are likely to explain the significant differences in quality of life between groups.
Previous histologic studies in animal models have shown that ablation lesion maturation is complete within 1 week after delivery (30,31). Thus, it seems unlikely that there is a significant mechanistic difference between reintervention after 2 months and reintervention at a later stage. Improvements in outcomes are apparently therefore primarily due to a difference in timing of reintervention, which pre-empts later recurrence.
Context of our results
PVI remains the cornerstone of successful AF ablation, with recent studies showing a lack of benefit from additional left atrial ablation even in persistent AF (3,4). However, the present study and others have shown how infrequently that is achieved with a single procedure (5,6). A recent study attempting to improve the durability of PVI by targeting a high force–time integral value still showed a significant rate of late reconnection and additionally reported an increased complication rate (32).
The relationship between PV reconnection and AF recurrence has been clearly established (7–10), and although it has previously been shown that complete durable PVI is not essential in all cases to prevent AT recurrence (33), nevertheless, the aim for operators at the initial procedure is to deliver exactly this goal. Furthermore, if a patient’s AF is truly PV trigger–driven, the only way to be certain that it will not recur is through complete durable PVI. Additional intervention to account for the limitations of the initial procedure is therefore a potentially logical adjunct.
Implications for clinical practice
Ideally, it would be possible to identify those patients with PV reconnection to focus re-intervention toward those with the most to gain from it but, at present, imaging modalities cannot accurately identify PV reconnection (34). We therefore tested a strategy of invasive re-assessment for PV reconnection in all patients and, although this approach has resulted in a clear overall improvement in outcomes, there would be significant health economic implications if this policy were to be adopted into clinical practice. However, the greatest benefit seems to have been in the subset of patients with early AT recurrence beyond 4 weeks within the conventional blanking period. We have previously reported that early recurrence in this time period is strongly associated with PV reconnection (35), and this and other studies have illustrated the relationship between such recurrence and post-blanking AT recurrence (25,36). Although there were no serious complications associated with elective repeat procedures in this study, we acknowledge that it was a relatively small sample size to detect these complications, and procedures were undertaken by experienced operators in a high-volume center. It may therefore be that targeting patients with such early recurrence could provide the optimal risk/benefit balance and cost-effectiveness.
First, the size of the study was small, having been powered to detect only a large absolute difference between groups. This approach was chosen deliberately because it was felt that the potential additional risk associated with a routine repeat procedure strategy could only be justified by a major improvement in success rates. However, although none of the repeat studies were associated with a complication, the size of this study means our ability to detect rare complications was limited. Second, this study assessed point-by-point ablation with radiofrequency energy performed by experienced operators. Its role in patients receiving alternate energy sources such as cryoablation cannot be assumed, and it is impossible to say with certainty whether our results would be replicated if performed by less experienced physicians. Third, we used validated handheld monitors to document AF recurrence, believing these provided the most comprehensive noninvasive rhythm monitoring option. Although patients were asked to provide recordings every day and whenever symptomatic, it is possible that asymptomatic episodes lasting <24 h may have been missed. Furthermore, follow-up was relatively short at 12 months. Finally, it was not possible to blind patients to treatment allocation due to the nature of the study, and this lack of blinding could have influenced quality-of-life scores.
A strategy of a routine repeat electrophysiology study to assess for and treat PV reconnection in patients with paroxysmal AF provided significant improvements in freedom from AT recurrence, AT burden, and quality of life compared with current standard care. Particular benefit from this strategy was seen in those with early AT recurrence beyond 4 weeks within the conventional blanking period.
COMPETENCY IN MEDICAL KNOWLEDGE: In patients with paroxysmal AF undergoing radiofrequency catheter ablation, routine repeat electrophysiology study after 2 months to assess for and treat pulmonary vein reconnection provided significant improvements in freedom from AT recurrence, AT burden, and quality of life compared with current SC.
COMPETENCY IN PATIENT CARE: Patients undergoing radiofrequency catheter ablation for paroxysmal AF should be made aware that there is only an approximately 40% likelihood that all PVs will remain isolated after the procedure. They should be informed that early re-intervention may be beneficial, especially if they experience early recurrence of AT beyond 4 weeks after the initial procedure.
TRANSLATIONAL OUTLOOK 1: Because this study was a relatively small, single-center trial, a larger multicenter study should be undertaken to assess the safety of a strategy of routine re-intervention in all patients.
TRANSLATIONAL OUTLOOK 2: Particular benefit from routine re-intervention to assess for and treat pulmonary vein reconnection was seen in those patients with early recurrence of AT in the blanking period beyond 4 weeks after the initial procedure. Further study of focusing this strategy on these individuals should be undertaken.
The PRESSURE Study investigators thank Dr. Tom Wong and Dr. Vias Markides from Royal Brompton Hospital for serving as the Data Monitoring and Safety Committee.
The PRESSURE study was supported by an investigator-initiated study grant from Biosense Webster, Inc. (IIS-239). The funders approved the study design but had no involvement in the collection, analysis, or interpretation of data; the writing of the manuscript; or the decision to submit for publication. Dr. Todd has received speaker fees from Boston Scientific and Medtronic. Dr. Hall has received speaker fees from Medtronic; speaker fees and fellowship support from Boston Scientific; and educational event support from Biosense Webster, Inc. Dr. Gupta has received speaker fees, research grants, and fellowship support from Biosense Webster, Inc.; and is on the St. Jude Medical Proctor's Panel for implantation of the Amulet device. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- atrial fibrillation
- Atrial Fibrillation Effect on Quality-of-Life Questionnaire
- atrial tachyarrhythmia
- pulmonary vein
- pulmonary vein isolation
- standard care
- Received October 31, 2016.
- Revision received January 17, 2017.
- Accepted January 19, 2017.
- 2017 American College of Cardiology Foundation
- Calkins H.,
- Kuck K.H.,
- Cappato R.,
- et al.
- Neuzil P.,
- Reddy V.Y.,
- Kautzner J.,
- et al.
- Kuck K.H.,
- Hoffmann B.A.,
- Ernst S.,
- et al.
- Verma A.,
- Kilicaslan F.,
- Pisano E.,
- et al.
- Nanthakumar K.,
- Plumb V.J.,
- Epstein A.E.,
- Veenhuyzen G.D.,
- Link D.,
- Kay G.N.
- Ouyang F.,
- Antz M.,
- Ernst S.,
- et al.
- Reynolds M.R.,
- Walczak J.,
- White S.A.,
- Cohen D.J.,
- Wilber D.J.
- Wokhlu A.,
- Monahan K.H.,
- Hodge D.O.,
- et al.
- Cappato R.,
- Calkins H.,
- Chen S.A.,
- et al.
- Cappato R.,
- Calkins H.,
- Chen S.A.,
- et al.
- Spertus J.,
- Dorian P.,
- Bubien R.,
- et al.
- Jais P.,
- Cauchemez B.,
- Macle L.,
- et al.
- Lin W.S.,
- Tai C.T.,
- Hsieh M.H.,
- et al.
- Wittkampf F.H.,
- Hauer R.N.,
- Robles de Medina E.O.
- Pratola C.,
- Baldo E.,
- Notarstefano P.,
- Toselli T.,
- Ferrari R.
- Harrison J.L.,
- Sohns C.,
- Linton N.W.,
- et al.
- Das M.,
- Wynn G.J.,
- Morgan M.,
- et al.