Atrial fibrillation: Catheter ablation

INTRODUCTION — The three principal goals of therapy in patients with atrial fibrillation (AF) are the alleviation of symptoms, the prevention of tachycardia-mediated cardiomyopathy, and the reduction in the risk of stroke. The first two goals can be achieved with either a rate or rhythm control strategy (see "Management of atrial fibrillation: Rhythm control versus rate control"). For patients in whom a rhythm control strategy is chosen, catheter ablation (CA) and antiarrhythmic drug therapy are the two principle therapeutic strategies to reduce the frequency or eliminate episodes of AF. (See "Maintenance of sinus rhythm in atrial fibrillation: Catheter ablation versus antiarrhythmic drug therapy".)

This topic will discuss the use of CA in patients with AF and provide the clinician with much of the information needed to discuss the procedure with the patient. The discussion of surgery to prevent recurrent AF is found elsewhere. (See "Atrial fibrillation: Surgical ablation".)

Stroke prevention is usually achieved with anticoagulation. This topic is discussed in detail separately. (See "Atrial fibrillation in adults: Use of oral anticoagulants".)

WHAT TO TELL YOUR PATIENT — When discussing CA to reduce symptoms in an AF patient, the following information should be provided:

●CA is a reasonable treatment option for AF patients when medications are unable to adequately control symptoms, are not tolerated, or not preferred.

●All patients who undergo CA must take oral anticoagulation for at least two to three months after the procedure. Anticoagulation should be continued long term in many patients with risk factors for stroke even if AF is not clinically present after the ablation. This is because patients may continue to have some AF episodes that may be asymptomatic; in addition, the reduction in AF burden seen post-ablation has not yet been shown to reduce stroke risk. It is a common misconception that patients who undergo successful ablation can stop oral anticoagulation.

●About 70 to 75 percent of patients are symptom free at one year [1]. A lower percentage is likely for persistent AF (about 60 percent).

●About 50 percent of patients have detectable AF at one year (this includes symptomatic and asymptomatic patients) [2,3].

●The success rate for ablation in patients with long-standing persistent AF (over one year) is poor.

●The risk of a major complication is about 4 percent, with vascular access complications being the most common. Other important, less common complications include stroke, cardiac perforation, or damage that includes injury to the pulmonary veins, esophagus, or phrenic nerve.

●The risk of dying within 30 days after an AF ablation procedure is about 1 in 1000 patients.

●The risk of a major complication is significantly higher at low-volume ablation centers. [4].

TECHNICAL CONSIDERATIONS — Technical considerations for CA are presented separately. (See "Catheter ablation for the treatment of atrial fibrillation: Periprocedural issues", section on 'Ablation techniques and targets'.)

COMPARISON WITH ANTIARRHYTHMIC THERAPY — For patients with symptomatic paroxysmal AF, either a trial of an antiarrhythmic drug or CA is a reasonable first-line approach. We are more inclined to perform CA in patients for whom the odds of success are high and if they prefer to avoid the use of long-term antiarrhythmic drug therapy. Studies comparing these strategies are discussed separately. (See "Maintenance of sinus rhythm in atrial fibrillation: Catheter ablation versus antiarrhythmic drug therapy", section on 'Patients without prior antiarrhythmic drug treatment'.)

EFFICACY — CA leads to significant symptom improvement in most patients. Over 70 to 75 percent are symptom free at one year. Some symptoms may be due to atrial or ventricular premature complexes rather than AF.

The absence of symptomatic AF recurrence is the primary efficacy outcome in most studies. However, with continuous invasive monitoring, approximately 50 percent of patients have had one or more documented episodes lasting 30 seconds or longer at one year. This becomes part of the rationale to continue long-term oral anticoagulation in many patients. (See "Catheter ablation to prevent recurrent atrial fibrillation: Anticoagulation", section on 'Our approach to anticoagulation'.)

How is recurrence defined and measured? — Recurrence of AF after CA is categorized as early or late. Each have distinct mechanisms and management implications [5]. From a clinical perspective, recurrences after the initial two-to-three-month post-ablation healing phase are more clinically relevant.

Early recurrences of AF are defined as those that occur within the first two to three months after CA. This period is often referred to as the "blanking period," and recurrences during this time are not included in studies examining the long-term success of AF ablation. Early recurrences occur as often as 40 percent of the time with radiofrequency ablation (RFA) [6] and about 17 percent of the time for those treated with cryoballoon [7]. It is postulated to be related to several potential mechanisms including sterile pericarditis, recovered pulmonary vein (PV) conduction, or proarrhythmic effects of the ablation procedure [8]. Some studies suggest that early recurrence appears to be a predictor of late recurrence, especially when the episodes occur late in the blanking period. However, most clinicians will treat early recurrences with antiarrhythmic drug therapy before consideration of repeat ablation in these patients. Episodes of AF occurring after three months are considered to be recurrent AF and are referred to as "late recurrent AF." The possible mechanisms for late recurrent AF following CA are discussed separately. (See "Mechanisms of atrial fibrillation", section on 'Specific clinical situations'.)

The frequency of late recurrent AF varies significantly across studies in part due to factors such as the method and intensity of surveillance, whether other atrial arrhythmias such as atrial flutter are counted, whether patients remained on antiarrhythmic drug therapy, and patient characteristics (eg, paroxysmal or persistent AF). In some studies, success has been defined as the absence of recurrent AF or other atrial arrhythmias with or without antiarrhythmic drug therapy; a more rigorous definition requires the absence of AF >30 seconds in patients not taking antiarrhythmic drugs.

The following studies illustrate the rates of late recurrence [9]:

●The DISCERN AF study evaluated episodes of symptomatic and asymptomatic AF (as well as atrial flutter and atrial tachycardia) before and after the procedure in 50 patients (80 percent with paroxysmal AF), using an implantable cardiac monitor capable of recording all AF episodes [10]. The total atrial arrhythmia burden was significantly reduced by 86 percent from a mean of two hours per day per patient before to 0.3 hours per day after. The ratio of asymptomatic to symptomatic episodes increased significantly after ablation from 1.1 to 3.7. After 18 months and a mean of 1.4 ablations, 58 percent of patients were symptom free.

●A 2023 meta-analysis of 73 studies (67,159 patients) with outcomes at >5 years found that the freedom from atrial arrhythmia was around 51 percent [11]. With multiple procedures, the success rate was nearly 70 percent. Results were better for those with paroxysmal compared with persistent AF but comparable between radiofrequency and cryoballoon ablation. Limitations of the analysis include the possibility of publication bias, significant heterogeneity among individual publications, and disparities in postablation AF surveillance.

●In the MANTRA and RAAFT-2 randomized trials, which allowed for antiarrhythmic drug use after CA, freedom from AF at two years was 85 and 72 percent, respectively [12,13]. (See "Maintenance of sinus rhythm in atrial fibrillation: Catheter ablation versus antiarrhythmic drug therapy", section on 'Patients without prior antiarrhythmic drug treatment'.)

●The 2019 CIRCA-DOSE study randomly assigned 346 patients with drug-refractory, paroxysmal AF to contact-force-guided RFA or two differing duration protocols for cryoballoon ablation [1] (see 'Technical considerations' above). All patients received an implantable loop recorder, and they also received noninvasive surveillance. Follow-up was for 12 months.

The primary endpoint of one-year freedom from atrial tachyarrhythmia (symptomatic or asymptomatic) as detected by continuous rhythm monitoring was about 53 percent in the three groups. One-year freedom from symptomatic atrial tachyarrhythmia, defined by continuous monitoring, ranged between 73 and 79 percent (p = 0.87). AF burden was reduced by about 99 percent in the three groups (p = 0.36).

Predictors of recurrence — Recurrence is more likely in patients with underlying cardiovascular disease such as hypertension, complicated heart disease (including valvular heart disease), older age, persistent as opposed to paroxysmal AF, procedure performed at a low-volume center, untreated obstructive sleep apnea, obesity, increasing plasma B-type natriuretic peptide level, or left atrial (LA) dilation [8,14-20].

●LA dilation – The success rate of CA is low in patients with long-standing persistent AF and severe LA dilation (>5.5 cm).

LA dilation should be assessed by volume determination rather than linear measurements if possible [21]. One study has shown that an LA volume ≥130 cc, assessed by computed tomography, predicts a recurrence rate of >90 percent at one year [22].

●Other LA remodeling parameters – Greater atrial wall thickness, lipid composition, and epicardial fat volume on cardiac computed tomography also predict AF recurrence in observational studies, but low measurement reproducibility may limit their clinical use [23]. Among 732 patients undergoing CA, 270 had AF recurrence after seven months. Patients with AF recurrence had higher LA wall thickness (anterior wall 1.9 versus 1.7 mm), epicardial adipose volume (145 versus 129 mm³) and lower LA wall attenuation reflective of higher lipid composition (-69.1 versus -67.5 Hounsfield Units).

Patients with persistent atrial fibrillation — The majority of patients in the studies of CA presented above had paroxysmal AF. The efficacy of CA in patients with persistent AF is lower than in patients with paroxysmal AF [24]. Our threshold for recommending CA is higher for patients with persistent AF given the lower success rates. Also, we avoid the use of CA as first-line therapy in patients with persistent AF.

We believe CA is a reasonable choice for individuals with symptomatic persistent AF who either fail or cannot tolerate antiarrhythmic drug therapy or, in certain circumstances (ie, tachycardia-mediated cardiomyopathy), where there may be a benefit to maintaining sinus rhythm even in the absence of symptoms.

To improve outcomes, standard pulmonary vein isolation (PVI) with or without additional ablative lesions can be performed. The utility of these additional lesion sets has not been consistently demonstrated, and we therefore recommend standard PVI without the creation of additional lesions, including posterior wall isolation for the first ablation attempt in the majority of patients [25]. We consider additional lesions in patients with long-standing persistent AF or a markedly enlarged LA. In patients with persistent AF, one small randomized trial (VENUS) found that the addition of vein-of-Marshall ethanol (see "Mechanisms of atrial fibrillation", section on 'Role of premature atrial complex and other arrhythmia triggers') infusion to CA, compared with CA alone, increased the likelihood of remaining free of AF at 6 and 12 months [26]. Further study of this procedure is needed.

A 2014 systematic review and meta-analysis identified 46 randomized trials and observational studies of 3819 patients who underwent CA for persistent AF [27]. Compared with medical therapy, CA reduced the risk of recurrent AF (odds ratio 0.32, 95% CI 0.20-0.53). Various ablation strategies were employed in the studies, and the most efficacious combined isolation of the PVs with limited linear ablation (eg, roof ablation, mitral isthmus ablation) within the LA. The success rate after two procedures was about 60 percent in all groups. (See 'Comparison with antiarrhythmic therapy' above.)

The STAR AF II trial was published subsequently to the meta-analysis [2]. In this trial, 589 patients with persistent AF were randomly assigned in a 1:4:4 ratio to ablation with PVI alone, PVI plus ablation of electrograms showing complex fractionated activity, or PVI plus additional linear ablation across the LA roof and mitral valve isthmus. There was no significant difference in the rates of the primary endpoint of freedom from any documented recurrence of AF lasting longer than 30 seconds after a single ablation procedure at 18 months (59 versus 49 versus 46 percent, respectively). Although serious adverse events appeared to be lower in the PVI-alone group, there were too few events for this endpoint to achieve statistical significance.

Patients with concomitant atrial flutter — Atrial fibrillation and flutter often coexist in part due to their common risk factors. In many atrial flutter patients, AF is thought to be the inciting arrhythmia, and as much as 55 percent of patients who undergo ablation for typical atrial flutter are also found to have AF on long-term follow-up [28]. Some studies have shown that PV triggers play an important role in the development of flutter [29].

While ablation of the tricuspid annulus-inferior vena cava (TA-IVC) isthmus is a highly successful treatment option for atrial flutter, the ablation approach to the patient with concomitant AF and atrial flutter requires a more extensive approach and has been evaluated:

●In a study of 108 patients with both AF and typical atrial flutter, patients were randomly assigned to either a dual-ablative procedure (PVI and TA-IVC isthmus ablation, 49 patients) or PVI alone (59 patients) [30]. After ablation, the following observations were made:

•During the first eight weeks post-procedure, none of the dual-procedure patients and 32 patients treated with PVI alone developed atrial flutter and required cardioversion and/or antiarrhythmic drugs.

•After eight weeks, all antiarrhythmic drugs were discontinued. Only three patients treated with PVI alone had further recurrences of atrial flutter, which was successfully treated with TA-IVC ablation. Seven of the dual-procedure patients and six of those treated with PVI alone developed recurrent AF. Of these 13 patients (12 percent of the total group), 10 underwent successful repeat PVI, and three remained in sinus rhythm on antiarrhythmic drugs.

These findings suggest that AF initiated by PV triggers may be the precursor rather than the consequence of atrial flutter. This conclusion is consistent with the observation that atrial flutter often starts after a transitional rhythm of variable duration, usually AF [31,32]. (See "Electrocardiographic and electrophysiologic features of atrial flutter".)

Attempts to control all atrial arrhythmias in patients with atrial flutter by performing PVI alone or at the time of an atrial flutter ablation have been studied:

●In the Triple A trial, 60 patients with atrial flutter but no documented AF were randomized to receive antiarrhythmic drugs alone, ablation of the cavotricuspid isthmus (CTI), or PVI. The primary endpoint, defined as any recurrent atrial tachyarrhythmia, occurred in 82.4 percent of the drug-treated group, 60.9 percent in the CTI group, and 10 percent in the PVI group during a mean follow-up time of 1.42 years [29].

●In the PReVENT AF study [33], 50 patients with atrial flutter and no documented AF were randomized to CTI ablation alone or with concomitant PVI. More patients in the isthmus-ablation-only group experienced new-onset AF during follow-up (52 versus 12 percent), and the one-year burden also favored the combined ablation group compared with the isthmus-ablation-only group (8.3 versus 4 percent).

These findings suggest that PVI either alone or in conjunction with atrial flutter ablation may have a beneficial effect on long-term suppression of all atrial arrhythmias. However, we do not recommend performing this procedure in lieu of or at the time of TA-IVC ablation in patients whose only documented arrhythmia is atrial flutter given the potential risks associated with additional ablation. (See "Atrial flutter: Maintenance of sinus rhythm".)

Patients with structural heart disease — The presence of structural heart disease may influence both the safety and efficacy of ablation procedures.

●Heart failure – CA appears to be safe and effective for the prevention of AF recurrence in patients with heart failure or impaired left ventricular function. The experience with ablation in this setting is discussed elsewhere. (See "The management of atrial fibrillation in patients with heart failure", section on 'Rhythm control'.)

●Cardiac resynchronization therapy – CA as an alternative to cardiac resynchronization therapy with atrioventricular node ablation in patients with heart failure is discussed separately. (See "The management of atrial fibrillation in patients with heart failure", section on 'Preference for rhythm over rate control'.)

●Mitral valve prosthesis – A potential concern with CA in patients with a mitral valve prosthesis is injury to the valve. Furthermore, entrapment of the ablation catheter in a mechanical mitral valve, necessitating open-heart surgery, has been reported in patients undergoing left-sided ablation procedures. This issue was addressed in a report of 26 patients with mitral valve prostheses who were compared with a matched group of 52 patients without a mitral valve prosthesis [34]. The rate of maintenance of sinus rhythm was the same in the two groups, but the patients with a mitral valve prosthesis had longer fluoroscopy times with greater radiation exposure and a higher rate of post-ablation atrial tachycardia (23 versus 2 percent).

●Rheumatic heart disease – The role of CA for chronic AF in patients with rheumatic heart disease (RHD) is not well defined. One study that compared 62 patients with AF, RHD, and mild mitral stenosis, and 124 matched patients without mitral stenosis, showed lower success rates (32 versus 57 percent) in those with RHD [35]. Another study performed electrophysiologic mapping in 17 patients with mitral stenosis who had chronic AF and were converted to sinus rhythm after balloon valvulotomy [36]. An organized atrial arrhythmia, which degenerated into AF, was induced in all patients; the focus was most often near the coronary sinus ostium. RFA was successful in 13 patients and, after a mean follow-up of 32 weeks, 10 were still in sinus rhythm.

●Cardiac surgery – Either the Maze procedure or off-pump CA using an epicardial approach should be considered in patients with AF and an indication for open-heart surgery. These approaches are not generally recommended for patients without an indication for cardiac surgery, except in special circumstances, because of the mortality and morbidity associated with surgery. (See "Atrial fibrillation: Surgical ablation", section on 'Maze procedure'.)

Patients with hypertension — Renal sympathetic nerve denervation has been proposed as an adjunctive treatment to CA in hypertensive AF patients. We do not feel the available evidence supports its use in this setting.

The rationale for the adding renal nerve denervation to CA is that hypertension is a major risk factor for the development of AF and that many hypertensive AF patients have increased sympathetic tone. Renal nerve denervation has been evaluated as a treatment for hypertension, but its efficacy has not yet been established. (See "Treatment of resistant hypertension".)

The issue of whether renal nerve denervation, when added to CA, can further lower the rate of AF recurrence was evaluated in the ERADICATE-AF trial [37]. In this study, 302 hypertensive (paroxysmal) AF patients were randomly assigned to CA or CA plus renal nerve denervation. The primary endpoint of freedom from AF, atrial flutter, or tachycardia at 12 months occurred in 56.5 and 72.1 percent of the two groups, respectively (hazard ratio 0.57, 95% CI 0.38-0.85). There was no significant difference in the rate of procedural complications between the two groups. Although the use of renal denervation as adjunctive therapy to CA improved the primary outcome, the lack of a sham-control group, that is CA plus sham renal denervation, is a major limitation of this study.

COMPLICATIONS — The types and rates of complications that occur in patients undergoing CA vary from series to series (table 1 and table 2). The overall rate of major complications is about 4 percent, with vascular access complications being the most frequent [38]. There may be an increased rate of adverse effects with more extensive circumferential ablation [39-43]. A meta-analysis of 89 studies including 15,701 patients showed an overall and severe procedure-related complication rate of 4.5 and 2.4 percent, respectively [44]. (See "Catheter ablation for the treatment of atrial fibrillation: Periprocedural issues", section on 'Ablation techniques and targets'.)

There are conflicting reports as to whether the complication rate has fallen with time:

●An analysis of 93,801 CA procedures performed in community hospitals in the United States between 2000 and 2010 did not identify a trend toward lower mortality [45]. The majority (81 percent) of procedures were performed in low-volume hospitals by low-volume operators. The overall frequency of complications was 6.29 percent, and there was a small but nonsignificant rise with time.

●In a meta-analysis of 192 published studies, including 83,236 patients, there was a significant decrease in the acute complication rate from 2007 to 2012 compared with 2000 to 2006 (2.6 versus 4 percent; p = 0.003) [46].

●An analysis of 15,701 patients from 89 studies showed that reported complication rates were lower in studies from 2013 to 2017 compared with 2018 to 2022 (5.3 versus 3.8 percent) [44].

In these studies, cardiac complications accounted for at least 50 percent of all complications. Most [45,47], but not all [48], studies have suggested that advanced age and female sex are risk factors for complications. In addition, annual operator (<25 procedures) and hospital volume (<50 procedures) have been associated with adverse outcomes [45].

Complications reported in series of patients undergoing CA to prevent recurrent AF will be reviewed here. Other complications that might occur with any electrophysiology study, such as radiation exposure and valve, vascular, or myocardial injury, are discussed separately. (See "Overview of catheter ablation of cardiac arrhythmias", section on 'Complications'.)

Mortality — Early case series found a death rate of about 1 to 1.5 in every 1000 patients [49,50]. More recent studies suggest a mortality closer to 5 per 1000.

A recent meta-analysis of studies performed between 2013 and 2022 showed a mortality of around 0.06 percent. Leading causes of death include cardiac tamponade and stroke. Other causes included pneumonia, pulmonary vein (PV) perforation, and sepsis [49].

Pericarditis — Pericarditis may occur as an early response to myocardial and/or pericardial injury caused by the ablation procedure. In a series of 1540 patients who underwent CA for AF, acute pericarditis was diagnosed in 57 patients (3.7 percent) [51]. The median symptom onset was one day after the procedure, and mean duration of treatment (with nonsteroidal antiinflammatory drugs with or without colchicine) was seven days. Cardiac tamponade was identified in only four patients (7 percent of patients with pericarditis). Radiofrequency ablation (RFA; rather than cryoablation) and younger age were independent risk factors for acute pericarditis postablation.

A delayed onset post-cardiac injury syndrome has also been reported following ablation procedures. (See "Post-cardiac injury syndromes".)

Cardiac tamponade — Cardiac tamponade resulting from perforation is the most frequent serious complication of CA for AF, occurring in slightly more than 1 percent of procedures using radiofrequency [47,49,52], and it is the leading cause of death [49]. Tamponade results from either catheter perforation of an atrial or ventricular free wall, especially with overheating during energy delivery, or less frequently with transseptal puncture. Some cases of tamponade may be caused by pericarditis; this may explain some cases with delayed onset. In one study of delayed tamponade, the median duration was 10 days, with a range of several hours up to 30 days [49]. Pericardial effusion associated with PV isolation (PVI) was significantly less common in patients who underwent cryoballoon ablation (0.8 versus 2.1 percent) in one meta-analysis [53], but a second meta-analysis showed no difference [54]. The treatment of tamponade caused by CA is similar to that in other settings. (See "Pericardial effusion: Approach to management", section on 'Indications for pericardial fluid removal'.)

Catheter entrapment — Entrapment of the circular mapping (LASSO) catheter in the mitral valve apparatus is a rare complication that can require cardiac surgery to resolve. The estimated incidence of this complication ranges between 0.01 and 0.9 percent, and specific sites of transeptal puncture or catheter manipulation may predispose to this adverse event [55-57].

Pulmonary vein stenosis — PV stenosis is a potential complication of ablation near or within the PVs. The lesion is characterized by fibrosis and scarring of the PV; specific pathologic changes include intimal thickening, thrombus formation, endocardial contraction, and proliferation of elastic laminae [58]. The diagnosis may be delayed or missed entirely, as symptomatic patients may come to attention months after their initial ablation. In one series, symptoms developed 4 ± 3 months after the most recent ablation, and the average delay between the onset of symptoms and diagnosis was 4.4 ± 5.4 months. Symptoms of PV stenosis include dyspnea with exertion (or less often at rest), cough, chest pain, hemoptysis, and recurrent lung infections [59,60]. The mean onset of symptoms is two to five months after the procedure [59-61]. The intensity of symptoms may be directly related to the degree of obstruction and inversely related to the duration of time to develop the stenosis [62]. Incorrect diagnoses including pneumonia, bronchitis, or suspected malignancy are often considered and result in unnecessary testing, treatment, and delayed intervention. Delays in diagnosis and treatment may allow for progression of stenosis and irreversible intraparenchymal lung damage.

The reported rate of PV stenosis depends not only on the factors described above, but also on the definition of stenosis severity and the intensity of screening. Early reports cited rates as high as 38 percent, but more studies cite rates for severe stenosis as low as 1 to 3 percent [47,55,60,63]. A minority of diagnosed patients appear to develop symptoms [64,65]. The incidence of severe PV stenosis is between 0.32 and 3.4 percent, but the risk may be lower with cryoballoon compared with radiofrequency energy [66]. The rate of PV stenosis requiring intervention may be as low as 0.1 to 0.3 percent [52].

Diagnostic evaluation for PV stenosis should be performed in patients who develop respiratory symptoms after RF ablation (RFA). The joint Heart Rhythm Society/European Heart Rhythm Association/European Cardiac Arrhythmia Society expert consensus statement of catheter and surgical ablation of AF suggests computed tomography or magnetic resonance imaging (MRI) as the preferred tests in suspected cases [62]. A ventilation/perfusion lung scan can also be used to diagnose PV stenosis.

Stent placement is a more effective therapy for PV stenosis compared with balloon angioplasty [66]. It is associated with a significant and almost immediate improvement in symptoms and pulmonary blood flow [59-61]. In series of patients who underwent balloon angioplasty with or without stenting, in-segment or in-stent stenosis requiring repeat intervention developed in approximately 50 percent of patients [60,61]. The roles of either elective stenting or surgery are not well defined [62]. However, the incidence of PV stenosis has significantly decreased due to improvement in ablative techniques, especially with moving the ablation lesions toward the atrial side of the PV-atrial junction.

Periprocedural embolic events — Patients undergoing CA to prevent recurrent AF are at risk for embolic events before, during, and after the procedure. The incidence of clinical stroke or transient ischemic attack is between 0 and 2 percent [9]. The role of anticoagulant therapy in this setting is discussed in detail separately. (See "Catheter ablation to prevent recurrent atrial fibrillation: Anticoagulation".)

MRI-detected brain lesions and cognitive impairment — Stroke and transient ischemic attack are not the only neurologic sequelae of CA. Multiple MRI studies performed within 24 hours after RFA have demonstrated new cerebral lesions in 7 to 44 percent of asymptomatic patients [67-69]. However, in a study of 60 AF patients at relatively low risk for stroke who underwent CA, only one patient developed new asymptomatic lesions on MRI soon after the procedure [70]. These lesions were presumed secondary to microemboli [71].

Studies of the impact of these lesions on neurocognitive function have come to differing conclusions as to the significance of these lesions, as illustrated by the following studies:

●The prevalence of cognitive impairment after RFA was evaluated in a study of 150 patients: 60 undergoing ablation for paroxysmal AF, 30 for persistent AF, 30 for supraventricular tachycardia, and 30 matched AF patients awaiting RFA (the control group) [72]. All RFA patients received periprocedural enoxaparin, and most patients with AF had a CHADS₂ score of 0 or 1 (table 3). All patients underwent eight neuropsychological tests at baseline and at 2 and 90 days after RFA. The prevalence of neurocognitive dysfunction at day 90 was 13, 20, 3, and 0 percent, respectively.

●In a study of 37 patients with paroxysmal AF who underwent 41 ablation procedures, MRI performed within 48 hours showed new brain lesions in 41 percent of patients and 44 percent of procedures [69]. Follow-up MRI at six months found glial scar in about 12 percent of those with lesions. However, there was no decline of neurocognitive function on testing.

Vascular complications — Vascular complications are among the most common adverse events related to AF ablation, likely due to the number and size of intravascular sheaths and the need for anticoagulation both during and immediately following the procedure. These complications include hematoma at the sites of catheter insertion, pseudoaneurysm, arteriovenous fistula, or retroperitoneal bleeding. Pseudoaneurysm and arteriovenous fistulae rates of 0.53 and 0.43 percent, respectively, have been reported [52,55]. This risk can be significantly reduced by the use of vascular ultrasound, which was demonstrated, in one study of 689 patients, to reduce the risk of vascular access complications from 5.3 to 1.1 percent [73]. Conservative management alone is usually sufficient for large hematomas and retroperitoneal bleeding, though anticoagulation may need to be held, and transfusion may be necessary in those patients where the risks of such interventions are warranted. Echo-guided manual compression and percutaneous intervention are usually effective treatments of femoral pseudoaneurysms or arteriovenous fistula, but direct surgical intervention is sometimes required [74].

Atrial esophageal fistula — This is a potentially life-threatening medical emergency for which the exact mechanism is unknown. The overall incidence is 0.3 to 0.54 percent, and mortality is between 50 and 83 percent [75]. A more recent international registry of patients with AF-CA showed that atrial esophageal fistula was diagnosed in 0.038 percent who had ablation with radiofrequency energy and 0.0015 percent who had cryoballoon ablation [76]. Early recognition can be missed due to the low awareness of this rare complication. It is important for patients to be educated as to warning signs and to contact their AF ablation center should any suggestive symptoms develop. Clinical manifestations usually present one to four weeks post-ablation (range of 2 to 60 days), and the most common symptoms are fever, chest pain, and recurrent neurologic events from septic emboli. Chest computed tomography is the preferred diagnostic modality. Endoscopy with air insufflation should not be performed.

Arrhythmic complications — New reentrant circuits created by the ablation lesions can lead to atypical left atrial (LA) flutter. These circuits tend to develop around regions of LA scar and often involve the perimitral region. Due to anatomic variability and technical challenges, successful ablation is more difficult than that for typical right atrial flutter involving the isthmus of the inferior vena cava and tricuspid annulus. A significant percentage of LA flutter following PVI may also involve the musculature of the coronary sinus or the roof of the left atrium [77]. (See "Electrocardiographic and electrophysiologic features of atrial flutter".)

Typical atrial flutter may also occur after LA ablation due to alterations in activation patterns of the LA and may have an unusual electrocardiographic morphology.

LA flutter appears to be more common following circumferential (as opposed to segmental) PVI [42,77-80]. In a randomized comparison of circumferential and segmental PVI, LA flutter developed in 9 of the 50 patients undergoing circumferential PVI, and in 1 of the 50 patients in the segmental PVI group [42].

In addition, many of the recurrent LA arrhythmias following segmental PVI are focal atrial tachycardias, as opposed to macroreentrant flutter circuits, and are often successfully treated with repeat isolation of the PVs.

Other — Other complications with their respective incidences are summarized:

●Phrenic nerve injury (<1 percent) [52,55] in patients receiving RFA (and up to 6.3 percent in those receiving cryoablation). (See 'Comparison of radiofrequency and cryothermal ablation' below.)

●Periesophageal vagal injury (gastric hypomotility) [41,81].

●Acute coronary artery occlusion/injury (<1 percent) [62,82].

●Iatrogenic atrial septal defect after cryoballoon ablation without clinical consequence (20 percent) [83].

COMPARISON OF ABLATION ENERGY SOURCES

Energy sources — There are four US Food and Drug Administration (FDA)-approved energy sources for AF CA: radiofrequency energy, cryothermal energy in the form of cryoballoon, laser balloon, and microsecond high-voltage electrical pulses (pulsed field).

Comparison of radiofrequency and cryothermal ablation — The commonly used approved energy sources for CA are radiofrequency and cryothermal energy. The efficacy and safety associated with these two energy sources have been found to be similar in multiple studies [1,53,84-88].

The three major randomized trials comparing the two energy sources are as follows:

●In the FIRE AND ICE trial, 762 patients with symptomatic, drug-refractory, paroxysmal AF were randomly assigned to cryoballoon ablation or radiofrequency ablation (RFA) [89]. The primary efficacy endpoint was the first documented clinical failure (eg, recurrence of AF, occurrence of atrial flutter or atrial tachycardia, use of antiarrhythmic drugs, or repeat ablation) following a 90-day blanking period after the index ablation. Arrhythmia surveillance was noninvasive.

The mean duration of follow-up was 1.5 years. The primary efficacy endpoint was similar in both groups (34.6 versus 35.9 percent, respectively; hazard ratio 0.96, 95% CI 0.76-1.22).

●In the FreezeAF trial, 315 patients with paroxysmal AF were randomly assigned to RFA or cryoballoon ablation [85]. The primary endpoint of freedom from atrial arrhythmia with absence of persistent complications was similar in the two groups at 12 months (70.7 versus 73.6 percent). Arrhythmia surveillance was noninvasive.

●In the 2019 CIRCA-DOSE study, which is discussed above, the two energy sources led to similar efficacy outcomes. (See 'How is recurrence defined and measured?' above.)

Complications of cryoballoon ablation may differ somewhat from standard RFA. Pericardial effusions, tamponade, and atrioesophageal fistula have been reported less frequently in cryoballoon ablation. A recent international registry of AF-CA patients showed that atrial esophageal fistula was diagnosed in 0.038 percent with RFA and 0.0015 percent with cryoballoon ablation [76]. Non-AF atrial tachyarrhythmias have also been less frequently reported in long-term follow-up of cryoballoon ablation. However, phrenic nerve paralysis has been reported in up to 6.3 percent of 1349 procedures, significantly higher than seen with standard RFA. Resolution occurs acutely in most patients and in >90 percent within one year [84]. The use of larger balloons that prevent distal ablation and the assessment of diaphragmatic compound motor action potentials have lowered the rate of this complication. Recordings of diaphragmatic electromyograms during cryoballoon ablation for AF accurately predict phrenic nerve injury [90].

Comparison of pulsed field and thermal ablation — Pulsed field ablation delivers microsecond high-voltage electric pulses to cause irreversible electroporation of cell membranes, causing cell death [91]. A rationale for this approach is that it may enable more precisely targeted ablation of myocardial tissue, minimizing injury of adjacent tissue.

A single-blind trial randomly assigned 607 patients with drug-refractory paroxysmal AF to undergo pulsed field ablation or thermal ablation (with either RFA or cryoballoon ablation) [92]. Rates of treatment success (defined as freedom from a composite of initial procedural failure, documented atrial tachyarrhythmia after a three-month blanking period, antiarrhythmic drug use, cardioversion, or repeat ablation) were similar in the pulsed field ablation and thermal ablation groups (73.3 and 71.3 percent). Acute and chronic serious adverse event rates were similar the two groups (2.1 versus 1.9 percent).

Registry data and a literature review have identified no reports of pulsed field ablation-associated symptomatic pulmonary vein stenosis or esophageal complications (which are rare severe complications associated with thermal ablation) [93,94].

In December 2023 [95] and January 2024 [96], the US FDA approved pulsed field ablation systems to treat paroxysmal AF.

PREVENTION OF RECURRENCE — The following therapies have been evaluated for their ability to prevent late recurrent AF; only treatment of obstructive sleep apnea (OSA) seems to be beneficial:

Glucocorticoid therapy – We do not believe there is sufficient evidence to recommend the use of prophylactic glucocorticoid therapy.

Two observations raise the possibility that corticosteroid therapy might be useful for the prevention of early recurrence. Firstly, inflammation is associated with the development of AF, and systemic and local inflammatory responses may result from radiofrequency ablation (RFA) [97] (see "Epidemiology, risk factors, and prevention of atrial fibrillation", section on 'Inflammation and infection'). Secondly, glucocorticoid prophylaxis reduces the risk of the development of perioperative AF in patients undergoing coronary artery bypass graft surgery. (See "Atrial fibrillation and flutter after cardiac surgery", section on 'Ineffective or possibly effective therapies'.)

The possible benefit from prophylactic glucocorticoid therapy was evaluated in a study of 125 patients with paroxysmal AF who were randomly assigned to either three days of glucocorticoid therapy or placebo starting immediately after the procedure [98]. The rate of AF recurrence (primary endpoint) was significantly lower in the glucocorticoid group at one month (27 versus 49 percent), with most of the benefit occurring during the first three days (7 versus 31 percent).

●Treatment of OSA – OSA is a predictor of recurrent AF after RFA. Patients with OSA who undergo CA should be encouraged to be evaluated for treatment with continuous positive airway pressure [99,100]. (See "Clinical presentation and diagnosis of obstructive sleep apnea in adults".)

●Colchicine – Colchicine, another drug with antiinflammatory properties, has been shown to decrease the risk of postoperative AF after cardiac surgery, particularly in patients with post-pericardiotomy syndrome. However, pending additional studies showing benefit, we do not use prophylactic colchicine. (See "Post-cardiac injury syndromes", section on 'Prevention'.)

The potential ability of colchicine to reduce the incidence of early recurrent AF after pulmonary vein isolation was evaluated in a study of 206 individuals with paroxysmal AF who were randomly assigned to colchicine 0.5 mg twice daily or placebo beginning on the day of CA and continuing for three months [101]. After follow-up of about 15 months, there was a lower recurrence rate of AF in patients taking colchicine (31.1 versus 49.5 percent; odds ratio [OR] 0.46, 95% CI 0.26-0.81).

●Angiotensin inhibition – The data are mixed as to whether angiotensin converting enzyme inhibitors/angiotensin II receptor blockers reduce AF after CA procedures. This issue is discussed elsewhere. (See "ACE inhibitors, angiotensin receptor blockers, and atrial fibrillation", section on 'Catheter ablation of atrial fibrillation'.)

Periprocedural weight reduction – Some studies suggest that periprocedural weight reduction may be a helpful adjunct to CA.

•Pre-procedure weight reduction – In a retrospective study of 600 patients, weight reduction before CA was associated with reduced AF occurrence [102]. Freedom from AF was observed in 420 patients (70 percent) at 15 months. Percent weight loss during the year before CA independently predicted freedom from AF through the next 15 months (OR 1.17, 95% CI 1.11-1.23).

•Post-procedure weight reduction – The SORT-AF Study compared one-year AF burden in patients with obesity participating in a weight loss program versus usual care after CA [103]. The intervention group had a small reduction in weight loss (5 versus 1 kg in controls). AF burden (measured with implantable loop recorder) after ablation did not differ between the two groups (OR 1.14, 95% CI 0.37-3.6). However, a reduction in body mass index was associated with a decrease in AF recurrence in persistent compared with paroxysmal AF patients.

FOLLOW-UP — Surveillance for recurrence of atrial arrhythmias is important in patients who have undergone CA. We agree with the joint Heart Rhythm Society/European Heart Rhythm Association/European Cardiac Arrhythmia Society expert consensus statement of catheter and surgical ablation of AF [62], which recommends the following:

●First visit with electrophysiologist at a minimum of three months, and then every six months for at least two years.

●Electrocardiograms (ECGs) at all visits; symptomatic (eg, palpitations) patients should be evaluated with some form of event monitoring.

The optimal method for screening for episodes of AF after ablation is not known. In the above studies, late recurrent AF was detected by patient symptoms, serial ECGs, 24- to 48-hour Holter monitoring, and implantable cardiac monitor [10,14,104,105]. Rhythm transmitters were also used in the first few months [105]. With the exceptions of implantable cardiac monitor, preexisting dual-chamber pacemaker, or implantable cardioverter-defibrillator with AF detection capabilities, these methods may underestimate the incidence of recurrence due to sampling error [106]. In addition, as has been well demonstrated, patients with AF have a high rate of asymptomatic episodes. (See "Paroxysmal atrial fibrillation", section on 'Natural history' and 'Efficacy' above.)

MANAGEMENT OF RECURRENCE — Some patients with symptomatic AF after CA are candidates for a repeat procedure. The decision to do so is usually based on a patient's assessment of the potential benefit and risks. Other patients may choose a trial of antiarrhythmic drug therapy to reduce symptoms.

The most common reason for recurrence of paroxysmal AF is reconnection of previously ablated electrically active tissue. Based upon the recurrence rates of AF after ablation, many patients are candidates for repeat ablation. We tell our patients that the success rate is in the range of 50 to 85 percent for a single procedure based primarily on AF type and anatomy, and that about 20 percent of patients have at least a second procedure. Success rates after a second procedure can be as high as 90 percent. Some experts and patients have agreed to repeat the procedure a third time. Patients with a history of persistent AF have a lower success rate and are less often felt to be good candidates for repeat procedures.

The issue of whether patients with AF recurrence should undergo a repeat procedure or be placed on antiarrhythmic drug therapy was addressed in a study that randomly assigned 154 patients with symptomatic, paroxysmal AF recurrence to either repeat ablation or antiarrhythmic drugs [107]. During three-year follow-up, fewer patients in the repeat ablation group demonstrated AF progression, defined as an increase in AF burden >30 percent relative to baseline based on insertable cardiac monitor (also sometimes referred to as implantable cardiac monitor or implantable loop recorder) data or development of persistent AF (25 versus 79 percent; p<0.01). Despite limitations of this study, it supports our approach of offering a second ablation procedure to most patients.

CONTRAINDICATIONS — While there are few absolute contraindications, the risks and benefits of AF ablation should be carefully considered in each patient.

Contraindications to AF ablation include preexisting left atrial or left atrial appendage thrombus, inability to safely administer anticoagulation during or after the procedure, inability to tolerate sedation, patients with atrial septal defect closure devices in whom transseptal access cannot be performed, and those with interruption of the inferior vena cava. While not contraindicated, ablations performed on those with very long-standing persistent AF (ie, >2 years), severe mitral stenosis or regurgitation, or large left atria are expected to have lower success rates. (See 'Predictors of recurrence' above.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Atrial fibrillation" and "Society guideline links: Arrhythmias in adults".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, “The Basics” and “Beyond the Basics.” The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on “patient info” and the keyword(s) of interest.)

●Basics topics (see "Patient education: Atrial fibrillation (The Basics)")

●Beyond the Basics topics (see "Patient education: Atrial fibrillation (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Role of catheter ablation (CA) – CA for atrial fibrillation (AF) leads to symptom improvement in many patients. However, it has not convincingly been shown to decrease the risks of embolization (eg, stroke) or death. (See 'Efficacy' above.)

●Efficacy – Current techniques for CA should lead to one-year freedom from symptomatic AF while off antiarrhythmic drug therapy in about 75 to 90 percent of patients with drug-resistant paroxysmal AF and no significant structural heart disease. (See 'Efficacy' above.)

●Complications – Important complications of CA include pericarditis, cardiac tamponade, stroke, vascular trauma, and phrenic nerve palsy, and, rarely, death (table 2). Specific signs and symptoms can help identify complications (table 1) . (See 'Complications' above.)

●Recurrence – For patients who have recurrent AF after a first ablation, there are two reasonable management strategies: a clinical trial of an antiarrhythmic agent or proceeding directly to a second ablation. Patients may have a preference for one or the other. (See 'Management of recurrence' above.)