AF ablation: indications, technique and anticoagulation

Catheter ablation is one of the rhythm control options in ESC 2024, alongside cardioversion, AADs, endoscopic and hybrid ablation, and open surgery.^[1] The wider condition is covered in Atrial fibrillation; drug rhythm control in Antiarrhythmic drugs; rate control in AF rate-control drugs; and anticoagulation around cardioversion in Cardioversion and anticoagulation timing.

What AF ablation is

Think of the patient who keeps coming back in AF despite flecainide, or who cannot tolerate amiodarone. ESC 2024 describes rhythm control as therapies dedicated to restoring and maintaining sinus rhythm; these include cardioversion, antiarrhythmic drugs (AADs), percutaneous catheter ablation, endoscopic and hybrid ablation, and open surgical approaches.^[1]

Catheter ablation is the percutaneous option.^[1] ESC 2024 states that it prevents AF recurrences, reduces AF burden, and improves quality of life in symptomatic paroxysmal or persistent AF when the patient is intolerant of AADs or does not respond to them.^[1]

The 2024 EHRA/HRS/APHRS/LAHRS expert consensus statement on catheter and surgical ablation of AF is a consensus document, so its advice carries no guideline class.^[3] On this page it is quoted as consensus, and every Class or COR comes from a formal ESC or ACC/AHA row.

Classifying the patient and the procedure

Two classifications matter: the temporal pattern of the AF, which drives the indication rows, and the strategy and energy used for the ablation.^[1]

Temporal pattern of AF

^{[1] [3] [1] [1] [3]}

How often it is done and who does well

In Australia and New Zealand, the 2023 CSANZ expert position statement notes that catheter ablation for AF has increased exponentially in many developed countries, including Australia and New Zealand.^[6]

^{[2] [1]}

Outcome depends on who is selected.^[2][3] The consensus lists modifiable comorbidities, AF type and duration, LA size, and abnormal atrial substrate on ECG and cardiac imaging among pre-procedural factors associated with a higher risk of recurrence.^[3]

• AF type (consensus): despite variation in the type and intensity of post-ablation rhythm monitoring, recurrence is lower in paroxysmal than in persistent AF.^[3]
• Time from diagnosis to ablation (consensus): each year of increase raises the risk of recurrence by 20% after adjustment for baseline comorbidities and medications; a meta-analysis of 6 observational studies (4950 patients) found a lower recurrence rate (relative risk 0.73) when the interval was under 1 year.^[3]
• Left atrial size (consensus): an independent pre-procedural predictor of recurrence; LA volume is a more accurate indicator of LA size than linear measurements.^[3]
• Fibrosis (consensus): in the DECAAF study, the extent of pre-ablation fibrosis on late gadolinium enhancement MRI independently predicted arrhythmia recurrence, although the consensus notes these findings have not been widely reproduced or employed.^[3]
• ACC/AHA 2023: patients with minimal atrial enlargement have the best outcomes, whereas increased myocardial fibrosis and more persistent forms of AF are associated with higher rates of recurrence after ablation.^[2]
• Age (ACC/AHA 2023): younger patients are likely to derive greater long-term benefit, including delaying AF progression.^[2]

Why isolating the veins works

The consensus states that paroxysms of AF are commonly triggered by ectopic beats from the PVs.^[3] It calls PVI a highly effective procedure in paroxysmal AF, in whom spontaneous PV firing is frequently the only trigger for AF paroxysms.^[3] It also calls PVI the cornerstone of AF ablation, required during all AF ablation procedures.^[3]

The consensus says the reported very low rates of persistent PV–LA conduction in the presence of entrance block, with contemporary ablation technology, indicate that entrance block alone is an adequate procedural endpoint during PVI.^[3] It adds that documenting exit block may prove useful when entrance block is ambiguous.^[3]

Going beyond the veins has not paid off as a routine strategy.^[2] ACC/AHA 2023 states that additional ablation targets beyond PVI as a routine strategy have not reduced AF recurrence or burden in RCTs.^[2] It warns that targeting large areas of atrial tissue may have negative consequences, including a higher likelihood of complications such as atrio-oesophageal fistula, poor LA mechanics, atypical flutters and an increased risk of stroke.^[2]

Heart failure changes the picture.^[3] The consensus describes AF and heart failure (HF) as frequently coexisting and potentiating each other in a vicious circle: AF begets HF and HF begets AF.^[3] In CAMERA-MRI, as the consensus reports, 68 patients with persistent AF and non-ischaemic cardiomyopathy were randomised to catheter ablation or medical rate control.^[3] LVEF improved significantly more with ablation than with rate control, LVEF normalised in 58% after ablation, and the results were maintained during long-term follow-up.^[3] The consensus adds that the favourable impact of ablation in AF with impaired LV systolic function extends beyond rhythm outcome and may frequently result in LVEF improvement.^[3]

^[3]

The patient who comes to ablation

The usual candidate has symptoms.^[3] The consensus lists palpitations, dyspnoea, dizziness, fatigue, pre-syncope and syncope, and notes that symptom scales such as the EHRA score and the Canadian Cardiovascular Society Severity of Atrial Fibrillation (CCS-SAF) scale standardise their assessment.^[3] It also warns that symptom–rhythm correlation is low in persistent AF, especially with comorbidities such as HF and diabetes.^[3]

The "asymptomatic" patient deserves a second look.^[3] The consensus calls the exclusion of a pseudo-asymptomatic status a key issue in patients without symptoms who remain in AF.^[3] Up to 77% of these patients may experience subjective symptomatic improvement, improved functional class and lower brain natriuretic peptide levels after sinus rhythm is restored by electrical cardioversion.^[3] So in asymptomatic patients, especially younger ones, the consensus considers a cardioversion worth attempting to see whether symptoms improve and the patient moves into the symptomatic category.^[3]

What else could be going on

Before ablation, the question is whether something other than AF drives the symptoms, or whether another arrhythmia drives the AF.^[3]

• Another supraventricular tachycardia as the trigger: the consensus considers it reasonable, preferably in younger patients and only when the SVT is judged the main trigger of AF, to simplify the procedure to elimination of the SVT only.^[3]
• Other supraventricular arrhythmias in patients undergoing AF ablation: ACC/AHA 2023 finds ablation of additional clinically significant supraventricular arrhythmias can be useful to reduce the likelihood of future arrhythmia (COR 2a, LOE B-NR).^[2]
• Symptoms from comorbidity: the consensus notes that patient symptoms are not specific and may come from coexistent cardiovascular conditions or AF risk factors, and symptom–rhythm correlation is low in persistent AF.^[3]

After ablation, the differential changes: ESC 2024 says recurrent arrhythmias after any AF ablation may manifest as AF, but also as atrial tachycardia.^[1] ESC 2024 notes that atrial tachycardia can be equally or more symptomatic for the patient than the original AF.^[1]

Selecting the patient

Shared decision-making comes first.^[1] ESC 2024 recommends shared decision-making when considering catheter ablation for AF, taking into account procedural risks, likely benefits and risk factors for AF recurrence (Class I, Level C).^[1]

ESC 2024 says many patients will not be suitable for catheter ablation because of factors that reduce the likelihood of a positive response, such as left atrial dilatation.^[1] ACC/AHA 2023 says ablation should be avoided in some patients because it is unlikely to succeed, for example with overwhelming substrate or ongoing processes that perpetuate AF.^[2] Its examples include, but are not limited to, advanced infiltrative cardiomyopathies such as amyloid, severe mitral stenosis or regurgitation, and cor pulmonale.^[2]

Risk factor work should not become an indefinite delay.^[3] The consensus says ablation should not be deferred in obese or physically inactive patients who have started lifestyle interventions and are progressing towards their goals.^[3] In morbid obesity (BMI above 40 kg/m²), it asks for an individualised risk–benefit assessment because of a higher complication rate and lower long-term freedom from AF.^[3]

Investigations before the procedure

The key pre-procedural question is thrombus.^[1] ESC 2024 states that left atrial thrombus is a contraindication to catheter-based AF ablation because of the risk of dislodgement leading to ischaemic stroke.^[1]

^{[1] [3]}

ESC 2024 reports that LA thrombus is more prevalent with elevated stroke risk scores and in non-paroxysmal than paroxysmal AF.^[1] Cardiac amyloidosis, rheumatic heart disease and hypertrophic cardiomyopathy (HCM) carry increased risk of ischaemic stroke and intracardiac thrombus even with adequate anticoagulation.^[1] ESC 2024 says cardiac imaging before ablation should be considered in these high-risk groups regardless of preceding effective OAC.^[1]

Thrombus imaging before ablation

^{[1] [3]}

ESC 2024 notes that observational studies suggest low-risk patients may be managed without visualising the LAA, but no RCTs have been performed.^[1] Practice varies: ESC 2024 lists transoesophageal echocardiography (TOE), intracardiac echocardiography (ICE) and delayed-phase cardiac CT.^[1]

Thrombus imaging modalities

^[3]

Emergencies in the lab

Here, resuscitation means the complication that declares itself on the table. The consensus calls cardiac tamponade the most frequent potentially life-threatening complication of AF ablation, with a reported incidence of 0.4% to 1.3% in recent large surveys.^[3]

^[3]

If bleeding persists despite protamine, the consensus lists fresh frozen plasma for warfarin, idarucizumab for dabigatran and andexanet for factor Xa inhibitors.^[3] Tamponade might also be delayed and can occur any time from an hour to weeks after the procedure; the consensus cites an incidence of 0.2% for delayed tamponade in a worldwide survey.^[3]

Indications: who should be offered ablation

The guidelines grade ablation by AF type, by prior drug failure and by heart failure.^[1][2] Read each row with its population attached.

Indication rows side by side

^{[1] [2]}

ACC/AHA 2023 adds a cost-value statement: catheter ablation for symptomatic AF provides intermediate economic value compared with AAD therapy (LOE B-R).^[2]

The evidence behind first-line ablation

ESC 2024 reports that multiple RCTs support catheter ablation as a first-line approach in paroxysmal AF, with a similar risk of adverse events to initial AAD treatment.^[1] It adds that it is not clear whether first-line ablation is superior to drug therapy in persistent AF.^[1] The consensus says the value of first-line ablation in persistent AF has not been specifically evaluated.^[3]

• EARLY-AF and STOP AF First (ACC/AHA 2023 summary): randomised trials of first-line ablation vs AADs in paroxysmal AF; recurrent atrial arrhythmias occurred in 43% with ablation vs 68% with AADs at 1-year follow-up in EARLY-AF, and in 25% vs 55% in STOP AF First (both P below 0.001).^[2]
• Safety (ACC/AHA 2023): across MANTRA-AF, EARLY-AF and STOP AF First, procedural complications occurred in 2% to 5% of patients.^[2]
• Who was studied (ACC/AHA 2023): AF burden was low with either strategy in EARLY-AF (time in AF 0.6% with ablation vs 3.9% with AADs), and the average age in all 3 trials was 60 years or less.^[2]
• Progression (consensus): in the 3-year follow-up of EARLY-AF, first-line cryoballoon ablation in paroxysmal AF was associated with a significantly lower incidence of persistent AF than AAD therapy (HR 0.25; 95% CI 0.09–0.70) using continuous cardiac monitoring.^[3]
• Energy (consensus): a pooled analysis found first-line RF ablation gave significantly higher freedom from recurrence than AADs in AAD-naïve paroxysmal AF.^[3]

After drug failure

• STOP-AF (ACC/AHA 2023): in patients who had failed at least 1 AAD and were randomised to another AAD or catheter ablation, ablation had a 1-year treatment success rate of 70% vs 7% with another drug.^[2]
• ThermoCool (ACC/AHA 2023): in paroxysmal AF after failure of 1 AAD, 66% were free from recurrent arrhythmia at 9 months with ablation vs 16% with another AAD.^[2]
• CABANA (ACC/AHA 2023): with 80% of patients on an AAD and judged ablation candidates, ablation was associated with a nearly 50% reduction in recurrent AF (HR 0.52; 95% CI 0.45–0.60; P below 0.001).^[2]

Timing and early rhythm control

ACC/AHA 2023 finds rhythm control can be useful to reduce hospitalisations, stroke and mortality in patients with a recent diagnosis of AF (under 1 year) (COR 2a, LOE B-R).^[2] EAST-AFNET 4 tested early rhythm control.^[7]

^[7]

ESC 2024 notes that rhythm control in EAST-AFNET 4 was predominantly pursued with AADs (80% of patients in the intervention arm).^[1] ACC/AHA 2023 reports that in a post hoc analysis of EAST-AFNET 4 the benefit was independent of symptom status, and catheter ablation was used for rhythm control in approximately 20% of patients.^[2] On timing within a strategy, ESC 2024 reports a small RCT in patients with paroxysmal or persistent AF in which delaying ablation by 12 months on optimised medical therapy did not affect arrhythmia-free survival compared with ablation within 1 month.^[1]

Heart failure with reduced ejection fraction

Here the ESC and ACC/AHA grades differ, so read each row with its body, year and population.^[1][4][2]

HFrEF rows

^{[1] [2] [4]}

ESC 2026 HF is the newer ESC document among the guidelines checked for this topic.^[4] Its row applies to selected patients with symptomatic AF and HFrEF who meet all three criteria: high-burden AF, under 1 year of continuous persistent AF, and a clear cause–effect relationship between AF and HF.^[4] For that group, its Class IIa, Level C row is the newer ESC grading for AF ablation in HFrEF, and the two ESC 2024 AF rows are dated history.^[4][1] Outside that group, including patients without symptoms and those who do not meet all three criteria, ESC 2026 HF has no AF catheter ablation row, and it has no separate row on tachycardia-induced cardiomyopathy.^[4] For patients outside that group, the ESC 2024 AF rows are the only ESC AF ablation rows for HFrEF among the guidelines checked for this topic, so read them with their year.^[1][4] ESC 2024 recommends AF catheter ablation in AF and HFrEF with a high probability of tachycardia-induced cardiomyopathy to reverse left ventricular dysfunction (Class I, Level B).^[1] It says ablation should be considered in selected AF patients with HFrEF to reduce HF hospitalisation and prolong survival (Class IIa, Level B).^[1] ESC 2026 HF says RCTs comparing ablation with pharmacological rhythm control in HFrEF cannot be considered conclusive for hard outcomes on HF hospitalisation and death, although ablation has proven superior to antiarrhythmic treatment for AF recurrence.^[4] It states that, based on current evidence, ablation should be considered only in carefully selected patients with HFrEF and symptomatic AF; this newer statement also bears on patients without symptoms, so read the 2024 rows for them alongside it.^[4]

Two of the HFrEF trials the consensus reports:^[3]

• CASTLE-AF: paroxysmal or persistent AF with HF (NYHA class II or above, LVEF below 35%) and an implantable cardioverter-defibrillator, unresponsive, intolerant or unwilling to take AADs; randomised to ablation or medical treatment with rate or rhythm control. Fewer patients in the ablation group had death from any cause or hospitalisation for worsening HF at 3.2 years (28.5% vs 44.6%; HR 0.62; P = 0.007), and mortality was also significantly lower with ablation (13.4% vs 25.0%; HR 0.53; P = 0.01).^[3]
• CASTLE-AF limits (consensus): sample size, strict selection criteria, generalisability, lack of blinded randomisation and treatment allocation.^[3]
• CASTLE-HTx: in symptomatic AF with end-stage HFrEF referred for heart transplant evaluation, ablation plus optimal medical therapy reduced the composite of death, left ventricular assist device implantation or urgent heart transplantation vs medical therapy alone over a median 18 months (8% vs 30%; HR 0.24; P below 0.001).^[3]

The consensus says proper patient selection is crucial for maximising benefit from AF ablation in HFrEF.^[3] Its Table 7 lists nine characteristics associated with LVEF recovery after ablation in impaired LV systolic function.^[3]

^[3]

The consensus notes that the Antwerp score, built on wide QRS, known HF aetiology, severe atrial dilatation and paroxysmal AF, has been shown to predict LV systolic recovery after ablation in HF.^[3] In patients with AF-induced cardiomyopathy who have recovered LV function, ACC/AHA 2023 says long-term surveillance can be beneficial to detect recurrent AF, given the high risk of recurrent arrhythmia-induced cardiomyopathy (COR 2a, LOE B-NR).^[2]

Heart failure with preserved ejection fraction

ACC/AHA 2023: in appropriate patients with symptomatic AF and HFpEF with reasonable expectation of benefit, ablation can be useful to improve symptoms and quality of life (COR 2a, LOE B-NR).^[2] ESC 2024 says the prognostic value of ablation in HFpEF is less well established than for HFrEF, and ESC 2026 HF says data on its role in HFpEF are limited.^[1][4] The consensus reports a prespecified CABANA subanalysis in patients with baseline HF symptoms, in which ablation reduced mortality by 60% vs drug therapy in the HFpEF subgroup (LVEF 50% or more).^[3]

The procedure: PVI and energy sources

ACC/AHA 2023 recommends PVI as the primary lesion set for all patients undergoing AF ablation unless a different specific trigger is identified (COR 1, LOE A).^[2] It rates the value of other endpoints beyond PVI, such as non-inducibility, and additional anatomical targets (posterior wall, low-voltage areas, complex fractionated electrograms, rotors) as uncertain (COR 2b, LOE B-R).^[2] ESC 2024 says the optimal ablation strategy has not been clarified in non-paroxysmal AF.^[1]

ACC/AHA 2023 reports a meta-analysis of 6 RCTs in which strategies including PVI were associated with a 50% reduction in recurrent AF compared with strategies without PVI.^[2]

Energy sources as printed

^{[3] [14]}

• Cryoballoon vs RF (FIRE AND ICE, as the consensus reports): 762 patients with drug-refractory paroxysmal AF randomised in a prospective multicentre design; cryoballoon was non-inferior to RF for efficacy and safety.^[3]
• Cryoballoon and the phrenic nerve (FIRE AND ICE): phrenic nerve injury was the most commonly reported complication at discharge in the cryoballoon group (2.7%); permanent palsy occurred in 0.3%.^[3]
• PFA vs thermal (ADVENT, as the consensus reports): 607 patients with drug-refractory paroxysmal AF were randomised to PFA or thermal ablation; at 12 months PFA was non-inferior for efficacy and safety.^[3]
• ESC 2024 on PFA: in a single-blind RCT of 607 patients, PFA was non-inferior for efficacy and safety endpoints compared with conventional RF or cryoballoon ablation.^[1]

The consensus says complication rates between RF and cryoablation do not seem to differ significantly, although the types differ.^[3] Persistent phrenic nerve palsy after PVI is seen almost exclusively after cryoablation, while oesophageal perforation is, in the vast majority, a consequence of RF.^[3] The 2024 consensus said the data for PFA were still limited, but that the existing evidence indicated an overall complication rate similar to the other two energy sources.^[3] A newer document is the 2026 scientific statement on PFA from EHRA, HRS, APHRS, LAHRS and the Canadian Heart Rhythm Society.^[14] It says randomised controlled trials and large registries have confirmed PFA efficacy at least equivalent to thermal ablation, with generally shorter procedure times and lower complication rates, particularly oesophageal and phrenic nerve injury.^[14]

Anticoagulation before, during and after ablation

ESC 2024 says any rhythm control procedure has an inherent risk of thromboembolism.^[1] ACC/AHA 2023 puts the risk of stroke in the first 30 days after ablation at 0.8%.^[2]

Peri-procedural OAC rows

^{[1] [2] [2] [1] [3]}

Uninterrupted or minimally interrupted?

ESC 2024 reports more silent stroke on brain MRI with interrupted OAC than with uninterrupted OAC.^[1] It also reports that randomised trials show comparable safety and efficacy for minimally interrupted OAC (withholding the morning DOAC dose on the day of the procedure) and a totally uninterrupted strategy.^[1] For once-daily DOACs taken truly uninterrupted, ESC 2024 says a pre-procedural shift to evening intake might be considered to mitigate bleeding risk.^[1]

• COMPARE (consensus): a large randomised trial in which ablation on uninterrupted vitamin K antagonist (VKA), compared with VKA interruption and low-molecular-weight heparin bridging, gave a significantly lower incidence of thromboembolic events and was associated with a lower minor bleeding risk.^[3]
• DOAC trials (consensus): VENTURE-AF, RE-CIRCUIT, AXAFA-AFNET 5 and ELIMINATE-AF compared uninterrupted rivaroxaban, dabigatran, apixaban and edoxaban with uninterrupted warfarin.^[3]
• Bleeding (consensus): meta-analyses show a 50%–55% relative risk reduction in major bleeding with uninterrupted DOAC vs uninterrupted warfarin at the time of ablation.^[3]
• Minimal interruption (consensus): the supporting RCTs were mostly single-centre, mainly in Asian populations, and too small to document non-inferiority.^[3]

In the lab

ACC/AHA 2023 says all patients undergoing catheter ablation require intraprocedural intravenous anticoagulation with heparin, or direct thrombin inhibitors in those with heparin allergies.^[2] The consensus reports a meta-analysis of 19 studies (7150 patients) in which patients with an activated clotting time (ACT) above 300 s had a significantly reduced risk of thromboembolic complications, without more bleeding, compared with ACT below 300 s, whatever oral anticoagulant was used.^[3] It adds that evidence supports giving the initial heparin bolus before transseptal puncture, and that patients on uninterrupted DOACs needed more UFH to reach target ACT, although not all DOACs interact with UFH in the same way.^[3] RE-CIRCUIT, as the consensus lists it, compared uninterrupted dabigatran with uninterrupted warfarin.^[3] The consensus reports a post hoc RE-CIRCUIT analysis in which patients on dabigatran needed similar amounts of UFH to VKA to reach therapeutic ACTs; other studies indicated that more UFH was needed with factor Xa antagonists.^[3]

Starting or restarting OAC after the procedure

• ESC 2024: post-ablation DOACs should be continued as per the dosing regimen when haemostasis has been achieved.^[1]
• Consensus: in patients not previously anticoagulated, DOACs are preferred over VKA after ablation because their immediate effect needs no bridging with UFH or LMWH.^[3]
• Consensus: if the patient was not anticoagulated or missed the last DOAC dose before the procedure, giving the DOAC 3–5 h after sheath removal is advisable, provided there is no evidence of mechanical complications.^[3]
• Consensus: where lifelong VKA is needed (for example a mechanical heart valve or rheumatic heart disease), ablation should be performed on uninterrupted VKA.^[3]

OAC after a successful ablation

The patient's question is simple: can the blood thinner stop now that the AF has gone? The consensus says management of anticoagulation beyond the early post-procedural period remains controversial.^[3] Both guidelines answer by stroke risk.^[1][2] ESC 2024 recommends continuing OAC according to the CHA2DS2-VA score and not the perceived success of the procedure (Class I, Level C).^[1] ACC/AHA 2023 says longer-term OAC should be dictated by stroke risk, such as a CHA2DS2-VASc score of 2 or more (COR 1, LOE B-NR).^[2]

To use that rule you need the general thresholds.^[1][2] ESC 2024 recommends a CHA2DS2-VA score of 2 or more as an indicator of elevated thromboembolic risk for decisions on initiating OAC (Class I, Level C), and says a score of 1 should be considered one (Class IIa, Level C).^[1] ACC/AHA 2023 recommends anticoagulation at an estimated annual thromboembolic risk of 2% or more per year, for example CHA2DS2-VASc 2 or more in men and 3 or more in women (COR 1, LOE A).^[2] It finds anticoagulation reasonable at 1% to under 2% per year, equivalent to CHA2DS2-VASc 1 in men and 2 in women (COR 2a, LOE A).^[2]

ACC/AHA 2023 explains why it continues OAC beyond 3 months by stroke risk.^[2] There were no randomised data that directly addressed the question, no consistent data showing a lower stroke risk after ablation, and higher stroke rates in people with CHA2DS2-VASc 2 or more who stopped OAC after 3 months.^[2]

• Denmark (ACC/AHA 2023): ACC/AHA 2023 calls the observational data on stroke after ablation mixed; in an observational analysis of 4050 patients with CHA2DS2-VASc 2 or more undergoing ablation in Denmark, stroke risk after ablation was low and comparable in those who did (0.93 per 100 patient-years) and did not (0.97 per 100 patient-years) discontinue OAC more than 3 months after ablation.^[2]
• United States (ACC/AHA 2023): a study of 6866 patients who underwent AF ablation in US clinical practice found that stroke risk was increased in patients with CHA2DS2-VASc 2 or more who discontinued OAC after 3 months (HR 2.48; 95% CI 1.11–5.52).^[2]
• Consensus summary of non-randomised meta-analyses: continued anticoagulation generally showed lower thromboembolic risk and favourable net clinical benefit with CHA2DS2-VASc 2 or more, and no significant benefit with CHA2DS2-VASc 1 or less.^[3]

Post-ablation OAC by stroke risk (2024 EHRA/HRS/APHRS/LAHRS consensus)

^{[3] [2] [1] [3]}

Trials published after the guidelines

This page summarises two randomised trials, published after the ESC 2024 and ACC/AHA 2023 rows were written, that test whether long-term OAC is still needed after successful ablation.^[8][9][1][2] Neither guideline row on this page was graded with them, so read them as new evidence beside the rows.^[1][2]

^{[8] [9]}

Rhythm drugs, recurrence and repeat ablation

For drug choice and safety, see Antiarrhythmic drugs. ESC 2024 reports that a short course of AADs (2–3 months) after ablation reduces early recurrences of AF, but does not affect late recurrences or 1-year clinical outcomes.^[1] ACC/AHA 2023 finds short-term AAD therapy after ablation can be useful in some patients to reduce early recurrences of atrial arrhythmia and hospitalisation (COR 2a, LOE A).^[2]

The consensus reports EAST-AF, which randomised 2038 patients (68% paroxysmal) to 3 months of AADs after ablation or no AAD.^[3] More patients on AADs were free of atrial arrhythmia during the 3-month blanking period, but freedom at 1 year did not differ (69.5% vs 67.8%; P = 0.38).^[3]

The blanking period

The consensus explains that atrial tachyarrhythmia can recur in the first weeks to months after ablation and some of these early recurrences may resolve with time.^[3] During the blanking period, a recurrence is not counted as treatment failure, and repeat ablation is usually not considered.^[3] With the current evidence, the consensus writing group recommends an 8-week blanking period after AF ablation.^[3] The 2026 PFA scientific statement describes that 8-week period as applying to thermal ablation modalities, shortened from 3 months; after PFA, it says a blanking period of just 1 month may be reasonable for clinical decision-making.^[14]

• Incidence (consensus): reported early recurrence ranges from 16% to 67%, depending heavily on the monitoring protocol.^[3]
• Freedom during blanking (consensus): patients free from recurrence during a 3-month blanking period had a 90% likelihood of remaining free at 12 months or longer.^[3]
• Timing matters (ADVICE, as the consensus reports): 401 patients with paroxysmal AF undergoing PVI were followed for 12 months with transtelephonic monitoring; 1-year freedom from AF was 77.2% without early recurrence, and 62.6%, 36.4% and 7.8% with early recurrence in the first, second and third month.^[3]
• ACC/AHA 2023: continued AADs beyond the blanking period, which it gives as 3 months, may reduce recurrent atrial arrhythmias out to 1 year.^[2]

After the blanking period, ESC 2024 frames recurrence management as a shared decision driven by the options for symptom control.^[1] It notes data supporting AAD continuation or re-initiation after ablation, even for previously ineffective drugs.^[1] ESC 2024 (Recommendation Table 19): repeat AF catheter ablation should be considered for AF recurrence after initial catheter ablation, provided symptoms improved after the initial PVI or after failed initial PVI, to reduce symptoms, recurrence and progression of AF (Class IIa, Level B).^[1] Its supporting text adds shared decision-making and clear goals of treatment.^[1] ACC/AHA 2023 warns that repeat ablations carry increased risks, rare but including PV stenosis and stiff LA syndrome.^[2]

Specific scenarios

Endoscopic and hybrid ablation

Endoscopic and hybrid ablation rows

^{[1] [2]}

AF ablation during cardiac surgery

• ESC 2024 (Recommendation Table 22): concomitant surgical ablation is recommended in patients undergoing mitral valve surgery with AF suitable for a rhythm control strategy, to prevent symptoms and recurrence of AF, with shared decision-making supported by an experienced team of electrophysiologists and arrhythmia surgeons (Class I, Level A).^[1]
• ESC 2024: in non-mitral valve cardiac surgery with AF suitable for rhythm control, it should be considered for the same aims, with the same team support (Class IIa, Level B).^[1]
• ACC/AHA 2023: for patients with AF undergoing cardiac surgery, concomitant surgical ablation can be beneficial to reduce the risk of recurrent AF (COR 2a, LOE B-R).^[2]
• ESC/EACTS 2025 valvular heart disease (newer ESC document): gives the same grades for valve surgery: concomitant surgical ablation is recommended with mitral valve surgery in AF suitable for a rhythm control strategy (Class I, Level A) and should be considered with non-mitral surgery (Class IIa, Level B), both to prevent symptoms and recurrence of AF, according to an experienced team of electrophysiologists and arrhythmia surgeons.^[12]

Left atrial appendage occlusion: pointers

LAA occlusion (LAAO) belongs here because it bears on stroke prevention after ablation.^[1][3] ESC 2024 describes percutaneous LAAO as a device-based therapy that aims to prevent ischaemic stroke in patients with AF.^[1]

LAA occlusion rows

^{[1] [2] [12]}

A newer US document, the 2025 SCAI/HRS clinical practice guideline on transcatheter LAAO, grades its advice with GRADE terms rather than COR and LOE.^[10] For patients with non-valvular AF (NVAF) and a contraindication to OAC, it suggests LAAO over no therapy (conditional recommendation, very low certainty evidence).^[10] For patients with NVAF who have decided to pursue stroke prevention, it suggests OAC or LAAO as treatment options; its corrigendum gives this as a conditional recommendation with moderate certainty evidence.^[10][11] After LAAO, it suggests either OAC or dual antiplatelet therapy (DAPT) (conditional recommendation, low certainty of evidence).^[10]

ESC 2024 says patients with a contraindication to all OAC options (the four DOACs and VKAs) have the most appropriate rationale for LAAO, despite post-procedure antithrombotic treatment that may carry a bleeding risk equivalent to DOACs.^[1] LAAOS III, as ESC 2024 reports it, randomised 4811 patients with AF to LAA occlusion or none during cardiac surgery for another indication; over a mean 3.8 years, ischaemic stroke or systemic embolism occurred in 4.8% vs 7.0% (HR 0.67; 95% CI 0.53–0.85).^[1] ESC 2024 notes LAAOS III did not compare occlusion with anticoagulation (77% continued OAC), and its supporting text accordingly presents surgical LAA closure as an adjunct to prevent thromboembolism in addition to anticoagulation.^[1] Its formal row recommends surgical LAA closure as an adjunct to OAC in patients with AF undergoing cardiac surgery (Class I, Level B).^[1]

After ablation in higher-risk patients who stop OAC, the consensus says LAA occlusion may be discussed as an alternative approach.^[3]

Complications and pitfalls

The consensus says major complications are usually defined as those that cause permanent injury or death, require intervention, or prolong or require hospitalisation.^[3] It puts the overall complication rate at 2.5%–8% and contemporary in-hospital death rates in experienced units as usually in the range of 0.05%–0.1%.^[3] ACC/AHA 2023 says complications occur in approximately 5% of patients, most of them vascular.^[2]

Complications after AF catheter ablation (ACC/AHA 2023 Table 26)

^{[2] [2]}

Atrio-oesophageal fistula

ACC/AHA 2023 calls an LA–oesophageal fistula the most severe complication: often fatal but quite rare.^[2] In the POTTER-AF registry of 553 729 procedures in 214 centres, as the consensus reports it, the incidence was 0.025%, and median times to symptom onset and diagnosis were 18 and 21 days.^[3] Its incidence was significantly higher with RF than with cryoballoon ablation (0.038% vs 0.0015%).^[3] The 2026 PFA scientific statement says that, to date, no case of oesophageal fistula has been reported despite PFA use in more than 500 000 patients worldwide.^[14]

Stroke and silent cerebral lesions

The consensus gives an incidence of stroke or TIA after ablation of 0.15%–0.5% in contemporary large series.^[3] Events typically occur within 24 hours, with the high-risk period extending through the first 2 weeks.^[3] It recommends a strict intraprocedural protocol: heparin (even before transseptal puncture), regular ACT measurement, an ACT of at least 300 s, and meticulous sheath management.^[3] Asymptomatic acute cerebral lesions on diffusion-weighted MRI may be seen in up to 30%, with no difference between VKA and DOACs.^[3]

Pulmonary vein stenosis

In large contemporary series, the consensus reports an incidence of severe PV stenosis of 0–0.5%; the highest risk is with RF ablation close to the PV orifices and/or within the PVs, significantly higher than with antral ablation.^[3] Symptoms usually appear weeks to months later: dyspnoea, haemoptysis, cough, pulmonary infections (which may be recurrent) or pneumonia, and chest pain.^[3] These may lead to misdiagnoses such as pneumonia, pulmonary embolism or even lung cancer.^[3]

• Grading (consensus): generally defined as mild under 50%, moderate 50%–70% and severe over 70% reduction in luminal diameter.^[3]
• Imaging (consensus): MRI or CT angiography is preferred.^[3]
• Treatment (consensus): interventional treatment is indicated for symptoms; asymptomatic or mildly symptomatic stenoses should be managed conservatively with watchful waiting.^[3]
• Restenosis (consensus): up to 50% even after acutely successful angioplasty; stenting is associated with a lower risk of restenosis than balloon angioplasty, particularly with larger-diameter and drug-eluting stents.^[3]

Phrenic nerve palsy

The consensus explains that the right phrenic nerve is most commonly affected because it runs close to ablation sites in the superior vena cava and both right PVs.^[3] Palsy occurs with all thermal technologies, but the vast majority of cases follow cryoablation.^[3] In a registry of 17 356 patients undergoing cryoballoon PVI, injury had recovered in 97.0% at 12 months, and 0.1% of the whole population had permanent injury.^[3]

• Presentation (consensus): can be asymptomatic but typically causes dyspnoea, tachypnoea, cough, hiccups and thoracic pain.^[3]
• Diagnosis (consensus): diaphragm excursion assessed by fluoroscopy (sniff test) or ultrasound.^[3]
• Prevention (consensus): monitoring the effects of right phrenic nerve pacing is now considered a standard part of cryoballoon ablation.^[3]
• Treatment (consensus): no active treatment is known to speed healing; in permanent palsy, diaphragmatic plication can improve dyspnoea and functional status.^[3]

Vascular and other complications

• Vascular (consensus): the most common major complications, including groin haematoma, femoral pseudoaneurysm, arteriovenous fistula and retroperitoneal bleeding, at 1%–4%; ultrasound guidance is recommended for vascular access.^[3]
• Coronary spasm with PFA (2024 consensus): it emphasised reports of severe spasm with the pentaspline PFA catheter, and said this can be mitigated by nitroglycerin before PFA applications at high-risk areas.^[3]
• Coronary spasm with PFA (2026 PFA scientific statement): proximity-related vasospasm was widely documented during cavotricuspid and mitral isthmus ablation, first with the pentaspline catheter and later similarly with several other PFA devices; nitroglycerine, as treatment or prophylaxis, can largely attenuate it.^[14]
• Haemolysis and acute kidney injury with PFA (2026 PFA scientific statement): post-PFA haemolysis and occasional acute kidney injury have been observed, correlated with a high number of applications, particularly with higher-voltage systems.^[14]
• Stiff LA syndrome (consensus): a rare complication, typically after multiple ablations, with unexplained dyspnoea and signs of right heart failure.^[3]
• Gastric hypomotility (consensus): from injury to the anterior vagal oesophageal plexus, usually with RF on the LA posterior wall; nausea, vomiting, bloating and abdominal pain within hours to weeks.^[3]

Prognosis and follow-up

Does ablation make people live longer? ESC 2024 says definitive evidence of prognostic benefit is needed before ablation can be considered for truly asymptomatic patients.^[1]

• CABANA (consensus): 2204 symptomatic patients aged 65 or older, or under 65 with 1 or more stroke risk factors, randomised to ablation or AAD/rate control; over a median 48.5 months, ablation did not significantly reduce the primary composite endpoint of death, disabling stroke, serious bleeding or cardiac arrest compared with medical therapy in the intention-to-treat analysis.^[3]
• CABANA limits: high crossover rates; the per-protocol analysis favoured ablation (P = 0.046) (consensus); ESC 2024 adds that low event rates may have diluted the effect.^[3][1]
• CABANA by age (consensus): in the CABANA trial, the clinical outcome of ablation vs AAD therapy varied with age, with the largest relative and absolute prognostic benefit in patients younger than 65 years, suggesting that selected subgroups may have clinical outcome benefit from ablation.^[3]
• Asymptomatic patients (consensus): the CABANA findings do not support ablation to improve prognosis in the general population of asymptomatic patients with AF.^[3]
• Older patients (consensus): a CABANA subanalysis found no prognostic benefit in patients 75 or older, with similar complications and recurrences.^[3]

ESC 2026 cardiac rehabilitation, a newer ESC document, says cardiac rehabilitation should be considered for patients with AF after radiofrequency catheter ablation to improve physical functioning (peak VO2, 6-minute walk distance) and health-related quality of life (Class IIa, Level B1).^[13]

ESC 2024 notes that intermittent rhythm monitoring has typically been used to detect relapses after ablation.^[1] ACC/AHA 2023 reminds us that recurrences are common: 30% to 40% after a first procedure in contemporary trials.^[2]

Special populations

• Older adults (ACC/AHA 2023): RCTs mainly enrolled patients younger than 70, who also had the largest benefits, but observational studies report improved quality of life with ablation in older patients.^[2]
• Older adults (consensus): two meta-analyses of observational studies found similar success with significantly more complications above 75 years.^[3]
• Hypertrophic cardiomyopathy (consensus): comparable efficacy for paroxysmal AF, poorer results in persistent AF, and early intervention before AF or substrate progression is of primary importance; major procedural complications appear more frequent.^[3]
• HCM and cardiac amyloidosis (ESC 2024): OAC is recommended in all patients with AF and either condition, regardless of CHA2DS2-VA score, to prevent ischaemic stroke and thromboembolism (Class I, Level B), which bears on post-ablation OAC.^[1]
• Women (consensus): in an observational cohort of 58 960 patients undergoing AF ablation (2016–2020), female gender was independently associated with a higher risk of hospitalisation longer than 1 day and with major and any adverse events.^[3]
• Obstructive sleep apnoea (consensus): up to 45% of patients referred for ablation have OSA, and there is no evidence that CPAP completely prevents recurrence or the need for ablation.^[3]

Evidence, guidelines and regional differences

^{[1] [2] [4]}

The 2024 EHRA/HRS/APHRS/LAHRS consensus adds procedural and follow-up advice that the guidelines leave open, such as the stroke-risk tiers for stopping OAC and the 8-week blanking period, without guideline classes.^[3]

In Australia and New Zealand

The 2018 NHFA/CSANZ AF guideline is the most recent Australian AF guideline among the guidelines checked for this topic (census 2026-10-09); it predates the ESC 2024 and ACC/AHA 2023 guidelines, so treat it as dated.^[5][1][2] Its published summary says failure of rate or rhythm control should prompt consideration of percutaneous or surgical ablation.^[5] The same summary recommends the sexless CHA2DS2-VA score: anticoagulation is not recommended for a score of 0 and is recommended for a score of 2 or more.^[5]

A 2023 CSANZ expert position statement on catheter and surgical ablation for AF, tailored to Australian and New Zealand conditions, is held only as its abstract, so none of its advice is used here.^[6]

Guidelines checked for this topic

Rows on this page come from the 2024 ESC AF guideline, the 2023 ACC/AHA/ACCP/HRS AF guideline, the 2026 ESC heart failure guideline, the 2025 ESC/EACTS valvular heart disease guideline, the 2026 ESC cardiac rehabilitation guideline and the 2025 SCAI/HRS LAAO guideline.^{[1][2][4][12][13][10]} Consensus advice comes from the 2024 EHRA/HRS/APHRS/LAHRS statement, and the newer PFA evidence from the 2026 EHRA/HRS/APHRS/LAHRS/CHRS scientific statement on PFA; nothing on this page takes a class from either.^[3][14] The ESC 2024 and ACC/AHA 2023 AF guidelines are the newest ESC and ACC/AHA AF guidelines among the guidelines checked for this topic (census 2026-10-09).

Exam pearls

• PVI is the cornerstone of AF ablation (consensus); ACC/AHA 2023 recommends it as the primary lesion set unless another specific trigger is found (COR 1, LOE A).^[2][3]
• First-line ablation within shared decision-making: ESC 2024 Class I, Level A in paroxysmal AF; Class IIb, Level C in selected persistent AF.^[1]
• HFrEF: ACC/AHA 2023 COR 1, LOE A in appropriate patients on GDMT with reasonable expectation of procedural benefit; ESC 2026 HF Class IIa, Level C in selected symptomatic patients meeting all three criteria.^[2][4]
• OAC: at least 3 weeks before (ESC 2024, elevated risk); uninterrupted during (ESC 2024), or continuous or minimally interrupted DOAC (ACC/AHA 2023); at least 2 months after in everyone (ESC 2024) or at least 3 months, longer by underlying risk (ACC/AHA 2023); longer-term OAC by stroke risk (ACC/AHA 2023) and continued according to CHA2DS2-VA rather than perceived success (ESC 2024).^[1][2]
• ACT of at least 300 s, regular ACT measurement and heparin even before transseptal puncture (consensus).^[3]
• Tamponade is the most frequent potentially life-threatening complication; hypotension in the lab is tamponade until proven otherwise (consensus).^[3]
• Atrio-oesophageal fistula: in the multinational POTTER-AF registry (553 729 ablation procedures in 214 centres) the median time from ablation to symptom onset was 18 days; the most common symptoms of oesophageal perforation are fever, chest pain or odynophagia, and neurological events; a barium swallow is contraindicated if a fistula is suspected (consensus).^[3]
• Persistent phrenic nerve palsy after PVI is almost exclusively a cryoablation complication; oesophageal perforation is, in the vast majority, an RF complication (consensus).^[3]

References14ShowHide

1. [1]Van Gelder IC, et al. 2024 ESC Guidelines for the management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS). Eur Heart J, 2024.PMID 39210723
2. [2]Joglar JA, et al. 2023 ACC/AHA/ACCP/HRS Guideline for the Diagnosis and Management of Atrial Fibrillation: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation, 2024.PMID 38033089
3. [3]Tzeis S, et al. 2024 European Heart Rhythm Association/Heart Rhythm Society/Asia Pacific Heart Rhythm Society/Latin American Heart Rhythm Society expert consensus statement on catheter and surgical ablation of atrial fibrillation. Europace, 2024.PMID 38587017
4. [4]Køber L, et al. 2026 ESC Guidelines for the management of heart failure. Eur Heart J, 2026.PMID 42661420
5. [5]Brieger D, et al. National Heart Foundation of Australia and Cardiac Society of Australia and New Zealand: Australian clinical guidelines for the diagnosis and management of atrial fibrillation 2018. Med J Aust, 2018.PMID 30067936
6. [6]Kistler PM, et al. 2023 Cardiac Society of Australia and New Zealand Expert Position Statement on Catheter and Surgical Ablation for Atrial Fibrillation. Heart Lung Circ, 2024.PMID 38702234
7. [7]Kirchhof P, et al. Early Rhythm-Control Therapy in Patients with Atrial Fibrillation. N Engl J Med, 2020.PMID 32865375
8. [8]Verma A, et al. Antithrombotic Therapy after Successful Catheter Ablation for Atrial Fibrillation. N Engl J Med, 2026.PMID 41211931
9. [9]Kim D, et al. Long-Term Anticoagulation Discontinuation After Catheter Ablation for Atrial Fibrillation: The ALONE-AF Randomized Clinical Trial. JAMA, 2025.PMID 40886309
10. [10]Goldsweig AM, et al. 2025 SCAI/HRS Clinical Practice Guidelines on Transcatheter Left Atrial Appendage Occlusion. J Soc Cardiovasc Angiogr Interv, 2025.PMID 41040442
11. [11]Goldsweig AM, et al. Corrigendum to '2025 SCAI/HRS Clinical Practice Guidelines on Transcatheter Left Atrial Appendage Occlusion': [Journal of the Society for Cardiovascular Angiography & Interventions. Volume 4, Issue 9 (2025) 103783]. J Soc Cardiovasc Angiogr Interv, 2026.PMID 41726062
12. [12]Praz F, et al. 2025 ESC/EACTS Guidelines for the management of valvular heart disease. Eur Heart J, 2025.PMID 40878295
13. [13]Bäck M, et al. 2026 ESC Guidelines on cardiac rehabilitation. Eur Heart J, 2026.PMID 42661418
14. [14]Kühne M, et al. Pulsed field ablation for the interventional treatment of atrial fibrillation: a scientific statement of the European Heart Rhythm Association of the European Society of Cardiology, the Heart Rhythm Society, the Asia Pacific Heart Rhythm Society, the Latin American Heart Rhythm Society, and the Canadian Heart Rhythm Society. Europace, 2026.PMID 41968953