Atrial fibrillation (AF) is the most common sustained arrhythmia worldwide, with an estimated prevalence of 2–4% in adulthood. With increased life expectancy, together with more accurate and intensive diagnostic methods, an estimated 2.3-fold rise is expected, with a lifetime risk of developing AF of one in three Europeans at the age of 55 years.1

AF is a complex disease in terms of management, with important morbidity and mortality, and has a significant impact on patients’ quality of life as well as a considerable socioeconomic burden. The techniques of AF ablation have evolved in the last three decades, and are currently dominated by percutaneous catheter ablation, which is available in the majority of hospitals, with a low periprocedural complication rate. As stated in the most recent (2020) European Society of Cardiology (ESC) guidelines,1 catheter ablation is a safe and superior treatment for symptom improvement and maintenance of sinus rhythm and a well validated treatment for the prevention of AF recurrence.

According to the latest data published by the European Heart Rhythm Association (EHRA)2,3 with information from 2016, AF ablation is one of the commonest electrophysiological procedures (reflecting a paradigm shift in the treatment of this arrhythmia), with substantial growth since 2007. Portugal is one of the top three southern European countries in AF ablation, with 478% growth and 82 ablations per million population (mean of 110 ablations per million population in ESC members). Unpublished official data presented by the Portuguese Association of Arrhythmology, Pacing and Electrophysiology (APAPE) show that 1633 AF ablations were performed in 2021 – a 35.2% increase from the previous year and corresponding to 157.86 ablations per million population.

A consensus statement on AF ablation published in 20174 laid out the periprocedural strategies, techniques, and endpoints for improving outcomes and safety with the use of ablation for AF treatment. However, there is little published information regarding the standards of practice for AF catheter ablation in the real world.

The purpose of this work was to collect continuous data on national activity regarding AF ablation, analyzing characteristics and indications among patients undergoing ablation and describing procedural characteristics and the frequency of associated complications. We aimed thereby to determine how clinical practice in Portuguese centers that perform AF ablation compares with the current scientific evidence and quality indicators for AF ablation as stated in the 2017 HRA/EHRA/ECAS/APHRS/SOLACE international consensus document4 and the 2015 Atrial Fibrillation Network/EHRA consensus conference.5

In recent years, AF has been diagnosed with greater frequency, due on one hand to increases in predisposing conditions and on the other hand to heightened awareness and earlier and systematic diagnosis.5 In our study, we analyzed a population of patients with previously diagnosed AF for whom a strategy of rhythm control with ablation was planned. The 337 patients included, reported by a third of all electrophysiological centers that perform AF ablation in Portugal, corresponded to 11.85% of all AF ablations performed in the two-year study period, reflecting a significant increase in the number of patients eligible for this rhythm control strategy in the country (with consistent increases in ablations performed of 15.3%, 37% and 26.8% from 2015 to date) and a surrogate of an increase in AF diagnosis.6,7

As AF is difficult to prevent, interventions for risk factor control to reduce incident AF should be increased, and there are significant measures that should be applied to improve care of patients diagnosed with the condition.8 The 2020 ESC guidelines for the diagnosis and management of AF1 propose a structured characterization of AF and an integrated approach to patient management. We identified a population of predominantly male (62.06%) AF patients with a median age of 65 years to whom ablation was offered, which is representative of the Portuguese population currently referred for ablation.6,7 This was a population with a high prevalence of risk factors for AF, including a significant proportion of overweight patients, with high rates of smoking, dyslipidemia and hypertension, which identifies a population at increased risk for whom an integrated care pathway should be offered with upstream therapy, to reduce recurrence. On the other hand, the proportion of patients with obstructive sleep apnea was low, probably reflecting the choice of medical treatment in this population with increased AF recurrence after ablation and for whom recurrences can be reduced with directed apnea treatment.1,4,8 While patients referred for ablation with moderate to severe structural cardiac disease was low (5.74% patients had LVEF<30% and only 4.77% had moderate to severe mitral regurgitation), the proportion of patients with LA dilatation was high (73.19%). This finding is in agreement with recent guidelines and published studies,9 which indicate that in highly symptomatic patients with structural cardiac disease there is a role for AF ablation, which can improve survival as well as symptoms (there is evidence for more robust efficacy if LVEF >25%). This is a population that on one hand could be more at risk from the procedure, but that on the other hand are likely to have a survival benefit and to whom ablation should be offered. Although LA diameter continues to be used to assess LA dilatation, in our population more robust volumetric measurements are increasingly used to assess LA structure before ablation. This issue should be addressed in further studies, to determine whether more accurate measurements of structural LA disease would help select patients who would benefit more from AF ablation.

With respect to AF characterization, most of the patients referred for ablation had paroxysmal AF (66.87%), while persistent and long-standing persistent AF (16.12% and 16.84%) were less frequent, which we believe reflects the current level of evidence for AF ablation in the different subsets of AF patients to whom the procedure is offered. The median mEHRA score was 3, with 92.55% of patients being symptomatic; additionally, previous failed attempts at rhythm control had been made in 85.59% of patients (24.53% with electrical cardioversion, 42.44% with class I antiarrhythmic therapy, 66.41% with class III antiarrhythmic therapy and 16.84% with previous AF ablation). These findings are in line with the most recent guidelines,1 which provide more robust evidence in symptomatic patients with paroxysmal AF and with failed previous attempts at rhythm control who should be offered AF ablation. However, patients with persistent and long-standing persistent highly symptomatic AF should also be offered a rhythm control strategy and ablation should be considered, and as such, they are also represented in our population (although in a less significant proportion).3,10,11

Integrated management of AF includes assessment of thromboembolic risk in all patients in order to identify those who should be offered oral anticoagulation.12 In the light of recent studies on DOACs,1 these should be the preferential choice in AF patients. In our cohort, the majority of patients were eligible for long-term anticoagulation (median CHADS2-VaSC2 score of 2) and a high proportion were offered anticoagulation (92.49%), of which 98.34% were with DOACs.

According to the AF ablation consensus statement,4 the procedure should be undertaken with sedation or general anesthesia, and in patients under anticoagulation who are in AF at the time of the procedure or in sinus rhythm without previous anticoagulation, performing preprocedural TEE is reasonable. In our cohort, 30.93% of patients underwent the procedure under general anesthesia and the other procedures were performed with conscious sedation. Although in patients under anticoagulation exclusion of LA appendage thrombi with imaging is not mandatory, in our population the majority of patients underwent cardiac CT (81.49%) or TEE (38.81%) before the procedure in order to exclude thrombus and also for procedural planning, as these modalities provide assessment of LA anatomy.

The consensus document recommends that ablation should preferably be performed without interrupting anticoagulation (class I recommendation),4 but regarding this, in our population there was a low proportion of patients (7.36%) with this indication. In fact, most procedures were performed according to a IIa class recommendation, with anticoagulation interrupted for 24 h (34.97%) or more than 48 h (49.39%). This is a quality measure that is unmet and efforts should be made to improve it.13 On the other hand, although anticoagulation was interrupted, these procedures were performed safely with the use of UFH (administered either before or immediately after transseptal puncture) and the use of ACT to monitor anticoagulation levels, which had an adequate median of 310 s during the procedure. Most patients restarted oral anticoagulation at a median of 5 h post procedure, as recommended, and 99.10% were anticoagulated at discharge (99.39% with DOACs). Although these data reflect a tendency to delay adoption of recommendations to perform ablation without interrupting anticoagulation (probably due to fear of complications), there are no concerns regarding the attainment of appropriate anticoagulation levels during all stages of the periablation period or after the procedure and on discharge, in a period with a higher thromboembolic risk due to extensive ablation in the left atrium. Of greater concern, although their number is low, is the fact that three patients (0.89%) were not anticoagulated at discharge, as after ablation and in the first two months it is necessary to maintain anticoagulation in all patients irrespective of their CHADS2-VaSC2 score.

In our study, the use of 3D mapping systems was predominant (76.52% of procedures), mostly using Carto® and Ensite® (59.15% and 16.77%, respectively). Although radiofrequency energy was most often used (73.87%), there was increasing use of single-shot techniques using cryoenergy, which represented a fourth of all procedures.14 This probably reflects a growing preference for techniques that are faster and enable more procedures to be performed, at a time when patient referrals for AF ablation are on the increase.

The aim of AF ablation should always include PVI, and although demonstration of bidirectional block is desirable, the class I recommendation refers only to the demonstration of at least entry block,4,5,15 demonstration of exit block being a class IIb recommendation. Concomitantly, in patients with a history of typical atrial flutter or its induction during ablation, cavotricuspid isthmus ablation is recommended. There is also the possibility of performing linear ablation as well as PVI, with pacing maneuvers to confirm block being recommended. In our study PVI was the strategy adopted in 94.61% of patients. It is important to note that 16.84% of patients were referred for a redo procedure after previous AF ablation, which is probably why PVI was not the main strategy in all patients. There was a high rate (98.17%) of acute PVI success during the procedures but entry block was confirmed in only 85.77%. Additional ablation lines were performed in only 12.05% of patients and pacing maneuvers to confirm block of these lines were performed in 39 of the 40 patients for whom this strategy was used. There were also 51 patients with typical atrial flutter previously diagnosed or occurring during AF ablation and cavotricuspid isthmus ablation was performed in 84.3%. As recommended, in our cohort the power delivered in the LA posterior wall was lower in all RF procedures, with an esophageal thermometer being used in only two patients; this a reasonable (class IIB) recommendation to reduce risk during AF ablation.4

It is recommended that outcomes of AF ablation should be reported to ensure that the most appropriate strategies are being used for patient management.4,16-18 In our study, we identified acute complications during AF ablation in five (1.49%) patients, of which only one was major, requiring intervention (emergency cardiac surgery). At one month post AF ablation, there were two reported deaths (0.73%) and six (2.18%) procedure-related complications. On patient also had an embolic event (under full-dose anticoagulant therapy) and two patients had a bleeding event (both while on anticoagulant therapy). No additional adverse events had occurred at one year after the procedure. This rate of reported complications is in line with previously reports, which may be an indication that adhering to recommendations for AF ablation increases procedure safety.

It is also recommended that patient recovery level and outcomes after a rhythm control strategy with ablation should be reported, along with AF management in the post-ablation period.17 In our study, at the end of the first month there was significant symptomatic improvement in the majority of patients (200; 86.21%) although 12.5% reported no change from baseline symptoms and 1.29% reported worsening of symptoms. The recovery level was sustained with no significant difference at one year post ablation, with a median mEHRA score of 1, a significant improvement compared to baseline (pre-ablation mEHRA score of 3).

Regarding patient management after ablation, at one month, six months and one year post ablation, 77.94%, 70.83% and 62.5%, respectively, of patients were still on Vaughan Williams class I or III antiarrhythmic drugs, reflecting a reduction in the use of this therapy compared to the pre-ablation period (85.29%). This result reflects good adherence to the class IIa recommendation4 for the immediate and short-term use of antiarrhythmic drugs in the post-ablation period and in the medium and long term, a strong preference for a hybrid approach (AF ablation and continued antiarrhythmic therapy), which is also a possibility for the care of patients with AF within the recommendations for rhythm control in the roadmap consensus for improvement of AF patient management.5

With respect to anticoagulation, 99.64% of patients were under anticoagulation therapy at two months post ablation and 93.33% by the end of the first year, which also reflects adherence to quality measures of patient management after ablation, for which maintenance of anticoagulant therapy should be driven by individual thromboembolic risk and not by apparent success of AF ablation.4

Regarding AF relapse, we identified 12.6% of patients who relapsed in the first month post ablation during the blanking period (four patients underwent a redo procedure), and at one year there were relapses in 19 (26.39%) patients (three underwent redo procedures). These findings are in line with reports of AF relapse after ablation,1 and probably reflect the importance of centers following recommended strategies in the selection of patients for ablation and also adhering to recommended strategies during the periablation period. In our study, 30-day AF relapse was independently associated with previous LA ablation and was less likely if PVI was the strategy used for ablation. In the long term (one year post ablation), AF relapse appears to be associated with increased LA diameter, AF type, lack of improvement in mEHRA score and the existence of AF relapse at 30 days. These results are in line with previous AF ablation trials and recommendations, with less evidence of long-term efficacy of ablation in non-paroxysmal forms of AF ablation, and suggest that relapse during the blanking period may predict long-term AF recurrence.1,9,10

More than 80% (86.7%) adherence to 13 of the 15 class I recommendations for AF ablation in the 2017 HRA/EHRA/ECAS/APHRS/SOLACE international consensus document4 was reported in patients in our study, and our results thus reflect good adherence to the consensus regarding patient management and selection of patients for AF ablation. Likewise, with respect to quality indicators specifically for AF ablation procedures and post-ablation management, there is a good standard of practice with respect to procedure planning, performance and management of patients in the post-ablation period.

Our study has limitations. First, this was not a mandatory registry, and participation in the survey increased the workload in participating centers. Consequently, although a significant proportion of hospitals responded to the survey, we do not have complete coverage throughout the country and there may be biases associated with differences between centers that did or did not respond to the survey. Second, as in all large-scale surveys, some data were missing, and additionally complete information was not mandatory, which could add bias to the reported results. In particular, one-year follow-up should be interpreted with caution due to the low number of patients for whom data are available, and further studies would be of great value to further address long-term follow-up. Also, the complications and deaths that occurred in the study period are not specified, and this is an important point that should be addressed in further studies.

In a population of patients with AF referred for ablation in Portuguese centers, we can state that patient management is provided according to the best scientific evidence recommended by European guidelines and that there is a high standard of practice with respect to the quality of AF ablation, with safety measures and reported outcomes achieved as recommended.

Pedro Carmo, Hospital de St.ª Cruz, Centro Hospitalar de Lisboa Ocidental E.P.E., Carnaxide, Portugal and Hospital da Luz, Lisbon; Dinis Mesquita, Hospital de S. Bernardo, Centro Hospitalar de Setúbal E.P.E., Setúbal, Portugal; Luis Brandão, Hospital Garcia de Orta E.P.E.; Luis Adão, Centro Hospitalar Universitário de São João, E.P.E.; Nuno Cabanelas, Hospital Prof. Doutor Fernando Fonseca E.P.E., Amadora, Portugal; Pedro Silva Cunha, Hospital de Santa Marta, Centro Hospitalar Universitário de Lisboa Central, E.P.E.; Sofia Almeida, Hospital CUF Tejo.

This study was sponsored by Boehringer Ingelheim Portugal in the form of an educational grant.

The authors have no conflicts of interest to declare.