Atrial fibrillation: Surgical ablation

INTRODUCTION — Atrial fibrillation (AF) is associated with an increased risk for stroke, heart failure, and death. An attempt to maintain sinus rhythm is made in patients based on the presence or absence of symptoms and evidence that myocardial function is being compromised. This may involve pharmacologic and nonpharmacologic strategies.

The most commonly performed invasive procedure used in an attempt to maintain sinus rhythm is catheter ablation performed by an electrophysiologist in a specially designed hospital procedure room. This technique is discussed in detail elsewhere. (See "Catheter ablation for the treatment of atrial fibrillation: Technical considerations for non-electrophysiologists", section on 'Ablation techniques and targets'.)

This chapter focuses on surgical ablation for the prevention of recurrent atrial fibrillation. The role of nonpharmacologic strategies for rate control in AF or to minimize thrombotic risk by left atrial appendage ligation, and an overview of the management of patients with AF, are presented separately. (See "Atrial fibrillation: Atrioventricular node ablation" and "Atrial fibrillation: Left atrial appendage occlusion" and "Atrial fibrillation: Overview and management of new-onset atrial fibrillation".)

RATIONALE — Work on the arrhythmic mechanisms for atrial fibrillation (AF) has led to a greater appreciation for the underlying process by which premature atrial complex (PAC; also referred to a premature atrial beat, premature supraventricular complex, or premature supraventricular beat) can promote areas of microreentry within the atrium that ultimately lead to AF. (See "Mechanisms of atrial fibrillation".)

Several surgical techniques have been developed to maintain sinus rhythm. In most cases, these techniques are employed in patients who are undergoing other cardiac surgery for some other reason (valve repair or replacement, coronary bypass grafting, or corrective surgery for congenital heart disease).

The three principal goals of surgical intervention include [1]:

●Creation of conduction block to disrupt all micro- and macroreentrant circuits. (See "Mechanisms of atrial fibrillation", section on 'Mechanisms of atrial fibrillation: triggers and substrates'.)

●Reestablishment or maintenance of electrical atrioventricular synchrony.

●Restoration of atrial mechanical function in order to improve diastolic filling.

MAZE PROCEDURE — Developed in the 1980s, the Maze procedure aims to create a "maze" of functional myocardium within the atrium that allows for propagation of atrial depolarization while reducing the likelihood that the wavefront would promote microreentry. (See "Mechanisms of atrial fibrillation".)

The most commonly performed procedure is referred to as the Cox-Maze IV and it consists of a pattern of linear scars created by incision or ablative technology such as radiofrequency or cryothermal ablation [2]. Most commonly, the procedure is performed at the time of other cardiac surgery, such as mitral valve surgery or coronary artery bypass graft surgery; the patient is on cardiopulmonary bypass. (See 'Indications' below.)

Traditional approach — The Maze procedure has evolved over the last 20 years. It originally created lines of scar by making several small incisions (referred to as "cut and sew") around the sinoatrial node as well as one to the atrial-superior vena caval junction (Maze I) through the sinus tachycardia region of the sinoatrial node. This unintentionally resulted in chronotropic incompetence and resulted in the Maze II procedure that modified the location of the incision to prevent this. The technical challenges of the Maze II procedure (eg, approach to the left atrium) eventually resulted in Maze III, which reduced the frequency of chronotropic incompetence, improved atrial transport function, and shortened procedure times (figure 1) [3-9]. Eventually, with the introduction of new technologies that created scar without incisions, such as cryothermia and radiofrequency, procedures were simplified and shortened. The "cut and sew" techniques became less frequently utilized and are now rarely performed.

Although they are more complex and slightly more time consuming, approaches that treat both atria appear to be more effective [10]. However, it is important to point out that long-term follow-up on the efficacy of these procedures is limited. The "gold standard" is considered to be the Cox-Maze IV procedure, in which lines of ablation are created using bipolar radiofrequency and/or cryothermal energy devices. Several lines of scar are created from the superior vena cava to the inferior vena cava on the right atrium, which is then connected to the tricuspid annulus. On the left, the posterior wall of the left atrium containing all four pulmonary veins is isolated as a box. The box is then connected to the mitral valve annulus. In addition, the left atrial appendage is removed.

The Maze procedure meets the three criteria for an ideal treatment of atrial fibrillation (AF) outlined above (see 'Rationale' above). In a five-year experience of 75 patients in one center, the procedure cured AF, restored atrioventricular synchrony, and preserved atrial transport function in all but one patient; six patients (9 percent) required antiarrhythmic medications [6]. Postoperative atrial pacemakers were implanted in 40 percent, mostly for preoperative sinus node dysfunction but occasionally for iatrogenic sinus node injury. This is a higher percent than has been observed in most contemporary series (see 'Need for pacemaker' below). Patients note a significant improvement in health-related quality of life, especially when compared with other cardiac procedures [8,11].

The addition of a Maze procedure does not appear to adversely impact outcomes after cardiac surgery. In a propensity-matched comparison of 485 patients undergoing Maze procedure and aortic valve replacement or coronary artery bypass graft surgery, there was no difference in any major morbidity or mortality. However, patients were on cardiopulmonary bypass longer and had a higher need for pacemakers [12]. Similar findings after mitral valve surgery have been reported in randomized trials [13].

A few studies have suggested that correcting AF at the time of cardiac surgery improves survival. In one study of 744 propensity-matched patients undergoing AF surgery at the time of other cardiac surgeries, patients who were successfully treated and free from antiarrhythmic medications at one year appeared to have improved survival, similar to patients without a history of AF. Those that remained in AF had a survival that was worse [14]. In another propensity-matched series of similar size, Maze-treated patients had a 10-year survival of 63 percent as compared with a 55 percent 10-year survival in similar patients whose AF was untreated [15].

Radial approach — The radial approach was developed after the Maze procedure to provide a more physiologic atrial activation-contraction sequence, thereby reducing the degree of left atrial dysfunction and optimizing the atrial contribution to left ventricular filling [16]. In contrast to the Maze procedure, in which the incisions desynchronize the activation sequence and often cut across the atrial coronary arteries, the incisions produced by the radial approach radiate from the sinus node toward the atrioventricular annular margins, paralleling the activation sequence and the atrial coronary arteries (figure 2A-B) [17].

Left versus biatrial lesion set — Some controversy exists as to the most appropriate lesion set. If only the left atrium is open in the case of a mitral valve repair or replacement without a tricuspid intervention, some surgeons have advocated a left-atrial-only approach [18]. In a series of 305 patients, there was an equal efficacy of a left-only approach, but a higher pacemaker implantation in those who underwent a biatrial lesion set [18]. Both of these findings were confirmed on a meta-analysis of 2225 patients from 10 studies [19]. The equivalent sinus restoration is supported by a randomized subgroup analysis of mitral valve patients that compared patients undergoing biatrial Mazes with pulmonary vein isolation alone [13]. However, in a 2008 analysis of 1723 patients, absence of a biatrial lesion set was a predictor of failure at 48 months [20]. This finding was also supported by a 2006 meta-analysis of 5885 patients in 69 studies [10]. More investigation may lead to a better understanding of which patients benefit from the biatrial lesion set.

One study compared the outcome of 23 patients who underwent a radial approach with 13 who had a traditional Maze procedure [21]. The radial approach was technically easier than the Maze, was equally likely to restore sinus rhythm (90 versus 92 percent for the Maze), and was associated with better left atrial transport function after surgery, as assessed with Doppler echocardiography. However, this approach has not yet been widely adopted by the majority of surgeons.

Minimally invasive approach — While the Maze procedure is usually performed at the time of cardiac surgery using a sternotomy, it has also been performed through a thoracotomy with a minimally invasive approach to mitral valves [22,23]. However, there are limited data available on the efficacy of using this approach.

Limitations and complications

Atrial transport function — One potential limitation to the Maze procedure is extensive damage to the atrial myocardium with resultant atrial dysfunction that may limit the hemodynamic benefit. In one series of 21 patients, the technique was effective in restoring sinus rhythm in 17 (85 percent) but atrial contractility improved in only 12 (71 percent) [24]. In one series of 150 patients who had successful return of electrical sinus, poor atrial contraction resulted in a significant rate of late stroke [25]. In a series of 31 patients, echocardiography demonstrated left atrial function in 71 percent and right atrial function in 81 percent; however, function was reduced compared with age-matched control subjects [26]. A successful Maze procedure may improve atrial transport, but the degree of improvement will likely be dependent on preoperative dysfunction.

Need for pacemaker — A common complication is the need for postoperative pacemaker, which usually occurs in approximately 10 percent of patients in most reports. As with simple electrical cardioversion, the mere restoration of sinus rhythm in patients with long-standing persistent AF can result in sinus node dysfunction unrelated to the surgery itself. Sinus node dysfunction may manifest as severe sinus bradycardia, sinus pauses or sinus arrest, sinoatrial exit block, atrial tachyarrhythmias, alternating periods of atrial bradyarrhythmias and tachyarrhythmias, and inappropriate heart rate responses during exercise or emotional stress. These arrhythmias probably result from partial denervation of the sympathetic and parasympathetic systems of the heart [27]. However, there may be some association with the extent of AF procedure performed. The published results are highly variable, with early studies reporting a postoperative need for pacemaker as high as 40 percent. Another study found pacemaker placement in 20 percent [6,13]. As surgeons learn to avoid the sinus node on the right side, the pacemaker rate may fall further. The manifestations of sinus node dysfunction after the modified Maze surgical technique are time dependent and, in one series, resolved within 12 months after surgery [28]; resolution correlated with functional reinnervation [27]. However, in another study, sick sinus syndrome developed in 7 of 87 patients (8.4 percent) and a pacemaker was required in 70 percent [29].

Pacemakers are rarely required after pulmonary vein isolation alone. In one series, there were more pacemakers required after biatrial lesion sets were employed, as compared with left atrial lesion sets alone [18]. This suggests that not all pacemakers are due to native sinus node dysfunction and that at least some are related to the creation of the right-sided lesions. There may be a balance between procedural efficacy and pacemaker requirement.

METHODS

Endocardial surgical ablation — Endocardial ablation is most often performed in an electrophysiologic suite by electrophysiologists using a catheter-based approach. However, the endocardial ablation of atrial tissue can also be performed in the operating room as an adjunct to cardiac surgery for other reasons or as a stand-alone procedure using a minimally invasive approach.

The efficacy of the left atrial approach was evaluated in a multicenter controlled trial in which 103 patients undergoing endocardial radiofrequency ablation during cardiac surgery for predominantly valvular heart disease were compared with 27 patients who refused radiofrequency ablation [30]. Linear ablation lesions were created around the right and left pulmonary vein ostia, with a lesion connecting the pulmonary vein lesions and another lesion connecting the left pulmonary veins to the mitral annulus. Patients treated with ablation were more likely to be in sinus rhythm at hospital discharge (63 versus 18 percent in the control group) and at one year (81 versus 11 percent).

A second study of 70 patients with either chronic or paroxysmal atrial fibrillation (AF) used minimally invasive surgical techniques with linear radiofrequency ablation to create a series of linear ablations around the orifices of the pulmonary veins in the left atrium [31]. During this procedure, the left atrium was accessed through small incisions in the chest (minimally invasive) and the patient was on femoro-femoral cardiopulmonary bypass. The mean intraoperative time was two hours, much shorter than with either the Maze procedure or catheter-based ablation techniques described below. At a mean follow-up of 1.5 years, AF was eliminated in 90 percent of patients.

Epicardial procedures — A minimally invasive approach using thoracoscopic pulmonary vein (and ganglionated plexi) isolation and ablation has been used to treat patients who have failed or are not candidates for antiarrhythmic drug therapy or catheter-based pulmonary vein catheter ablation (CA) [32-34]. However, we do not use this approach as the first interventional strategy.

As a stand-alone procedure, these approaches are limited to pulmonary vein isolation, either as a box or islands around each side. An endocardial approach is necessary to connect any lesion to either the mitral or tricuspid valves. (See "Atrial fibrillation: Catheter ablation".)

This procedure is performed with video-assisted thorascopic access to the epicardial space through small, either right-sided or bilateral thoracic incisions and primarily focuses on pulmonary vein isolation. The pulmonary veins are electrically isolated with a bipolar radiofrequency ablation clamp or suction-assisted unidirectional device. Depending on the center, additional ablation of ganglia or the left atrium are also performed epicardially.

Individuals are not placed on bypass and linear epicardial lesions are delivered to the pulmonary veins. This technique can be used in a stand-alone fashion or in conjunction with an endocardial approach (see 'Hybrid approach' below). Complications of the traditional endocardial approach to catheter ablation, such as damage to the phrenic nerve and the development of thromboembolism, occur but may be less frequent through an epicardial ablation.

The 2020 CASA-AF trial randomly assigned 120 patients with long-standing persistent atrial fibrillation (AF) to thoracoscopic surgical or catheter ablation [35]. The primary outcome of single-procedure freedom from AF/atrial tachycardia ≥30 seconds without antiarrhythmic drugs at 12 months occurred at a similar rate in both groups (26 versus 28 percent, respectively; odds ratio 1.13, 95% CI 0.46-2.83). One death was reported in the surgical ablation group, and over 12 months it was more expensive and provided fewer quality-adjusted life-years compared with catheter ablation (0.78 versus 0.85; p = 0.02).

The FAST trial randomly assigned 124 patients with antiarrhythmic drug-refractory AF with left atrial dilatation and hypertension (33 percent) or failed prior CA (67 percent) to either minimally invasive surgical ablation or CA [36]. At 12 months, the primary end point of freedom from left atrial arrhythmia of greater than 30 seconds without antiarrhythmic drugs was significantly higher in the surgical ablation group (65.6 versus 36.5 percent). However, there were significantly more periprocedural complications, such as pneumothorax, major bleeding, and the need for pacemaker, in the surgical ablation group (35.4 versus 15.9 percent). The need for pacemaker placement was 3 percent after surgery and 0 percent after catheter ablation.

Hybrid approach — Surgeons and cardiologists are now combining to perform some of the lesions surgically and some in the electrophysiology lab or operating room in a "hybrid" approach. There are three different types of approaches: right thoracotomy [37], subxyphoid [38], and bilateral thoracoscopic [39].

The convergent procedure is a subxyphoid approach that requires both surgery and catheter ablation in a single hospitalization to complete. Long-term follow-up is needed. Both the right and bilateral thoracoscopic approaches allow for a staged or concomitant approach. The right thoracotomy uses unipolar suction-assisted radiofrequency ablation. The bilateral uses bipolar radiofrequency ablation clamps and is the only approach that allows for treatment of the left atrial appendage. Although initially a simultaneous or single-stage approach predominated, there has been a trend toward a staged approach, performing the electrophysiology portion only in the patients that fail initial surgery. In a small single-center series, the staged, bilateral approach has been demonstrated to be as efficacious as a "cut-and-sew" open Maze, but none of the reports is large enough to make definitive recommendations for any of these procedures. However, they do offer the potential for improving the success of treatment in challenging patients without the full extent of surgery or its complications. They may be considered in medically refractory AF patients who are unlikely to be successfully treated with catheter ablation alone.

Comparison of methods — The relative efficacy and safety of the different procedures has not been well studied. There are no large randomized trials, and available observational data are inconsistent, as illustrated by the following:

●In a review of 70 patients, 40 underwent radiofrequency (RF) ablation and 30 underwent the surgical Maze procedure [40]. Those undergoing RF ablation and those undergoing the surgical Maze had similar rates of sinus rhythm at discharge (85 versus 73 percent) and at one year (91 versus 96 percent).

●A series of 56 patients who underwent RF ablation during surgery were compared with 56 matched historical controls [41]. Patients treated with the conventional surgical Maze were more likely than those treated with RF to be in sinus rhythm at discharge (88 versus 63 percent) and at last follow-up (92 versus 62 percent).

●In a series of 377 patients from a single institution, 220 underwent a surgical Maze procedure, and the remaining 157 were treated with RF ablation during surgery [42]. Patients treated with the conventional surgical procedure were more likely than those treated with RF to be in sinus rhythm at three months (91 versus 62 percent) and at six months (90 versus 75 percent).

However, these results should be interpreted with caution for the following reasons:

●Each cohort included a heterogeneous mix of patients undergoing different surgeries with a variety of underlying cardiac disease and comorbidities.

●The RF procedures were not standardized, and in most cases, documentation of effective RF lesions was not performed. In percutaneous ablation procedures, documentation that ablation produces conduction block is an important predictor of success. (See "Atrial fibrillation: Catheter ablation", section on 'Efficacy'.)

INDICATIONS — We believe it is reasonable to attempt to surgically ablate atrial fibrillation (AF) in carefully selected patients with a high burden of AF. Most of these patients will have been referred for cardiac surgery for another reason, such as significant valvular or coronary heart disease. It is uncommon for patients without an indication for open heart surgery to be referred for a surgical ablative procedure.

We believe it is reasonable to perform concurrent surgical ablation at the time of mitral valve surgery in patients with a high burden of AF, including those with paroxysmal or persistent AF [2]. In patients undergoing mitral valve surgery, most often for mitral regurgitation, 30 to 50 percent have AF [14,43]. Most studies of the surgical treatment of AF have enrolled patients undergoing mitral valve surgery, with the PRAGUE-12 randomized study being one exception [44]. The procedure should be performed only when it does not add significant additional surgical risk. Patients with prior cardiac surgery or those who cannot tolerate single lung ventilation are less than ideal candidates.

In patients who have failed or are intolerant to antiarrhythmic drug therapy, have failed catheter ablation in the electrophysiology laboratory, and in whom a minimally invasive approach is feasible, it is reasonable to attempt this procedure [45]. (See 'Minimally invasive approach' above.)

EFFICACY — In most published series that compared patients undergoing surgical treatment of atrial fibrillation (AF) with those not, the burden of AF was lower with surgical treatment. This generally leads to improved quality of life and a lower use of antiarrhythmic drug therapy. Similar to patients undergoing traditional catheter ablation, there is no evidence that survival is improved [13,44,46].

The best evidence of long-term outcomes comparing ablation with no ablation during mitral valve surgery comes from a study of 260 patients with persistent or long-standing persistent AF who were randomly assigned to either surgical ablation or no ablation during the surgery [13]. Patients who were assigned to ablation received either pulmonary vein isolation or a biatrial Maze procedure. All patients underwent closure of the left atrial appendage. The primary end point of freedom from AF at both 6 and 12 months, as assessed by three-day continuous monitoring, occurred more often with ablation (63.2 versus 29.4 percent; p<0.001). There was no significant difference between the two ablation procedures for this outcome. Mortality was 6.8 percent in the ablation group and 8.7 percent in the no ablation group (hazard ratio 0.76, 95% CI 0.32-1.84). There was no significant difference in the rate of a composite secondary outcome that included cardiac or cerebrovascular adverse events, nor were there differences in the end points of functional class, quality-of-life measures, and medication use. However, the rate of implantation of a permanent pacemaker was higher with ablation (21.5 versus 8.1 per 100 patient-years; p = 0.01). Individual patient factors, such as symptom status, should determine whether AF ablation is performed in this setting. It is possible that the restoration of sinus rhythm might lead to improved clinical outcomes with longer follow-up in this relatively small study. (See "Atrial fibrillation: Catheter ablation", section on 'Efficacy'.)

A number of observational studies have demonstrated that performing a combined procedure with either "cut and sew" Maze [47-52] or catheter ablation [20,30,53-56] leads to a 60 to 80 percent likelihood of freedom from AF at one year. In an observational study of 576 patients who underwent the Cox-Maze IV procedure, overall freedom from AF was 78 percent at five years and freedom from AF off antiarrhythmic drugs was 66 percent [2]. In this study, there was no difference between those with paroxysmal and those with persistent/longstanding AF. The freedom from AF after surgery in single-center series has been reported to be as high as 92, 84, and 77 percent at 1, 5, and 10 years, respectively [57].

TREATMENT FAILURES — Surgical approaches to prevent atrial fibrillation (AF) have a high rate of success (see 'Maze procedure' above). However, some patients have recurrent atrial arrhythmias including AF, typical flutter, atypical atrial flutter (often due to reentrant circuits around the surgical scars), and focal atrial tachycardias [58]. Such patients may be candidates for electrophysiology study and catheter ablation. (See "Atrial fibrillation: Catheter ablation", section on 'Management of recurrence' and "Atrial fibrillation: Catheter ablation", section on 'Efficacy'.)

ANTICOAGULATION

First two to three months — All patients are anticoagulated with a direct oral anticoagulant (DOAC) or vitamin K antagonist (VKA) for at least two months after surgical ablation, regardless of their CHA₂DS₂-VASc score or rhythm status.

Subsequent anticoagulation decisions — In patients who are at risk for stroke, anticoagulation therapy should be continued indefinitely after surgical ablation procedure regardless of the rhythm outcome [59]. Since surgical ablation to prevent AF can restore sinus rhythm long term, some have suggested that it may also reduce stroke risk and therefore the need for long-term anticoagulation, but there is no convincing evidence to support this approach. (See "Atrial fibrillation in adults: Use of oral anticoagulants".)

●Identifying patients at risk for stroke – Identifying which patients are at risk for stroke and require long-term anticoagulation is not different in patients who have undergone surgical ablation. This is discussed separately. (See "Atrial fibrillation in adults: Selection of candidates for anticoagulation".).

There is limited information regarding the validity of the CHADS2 stroke score in predicting stroke risk in surgical ablation patients. Observational studies suggest overall low stroke rates following surgical ablation [60-63]. In a Swedish cohort study, 526 patients had a Cox-maze III procedure, including left atrial appendage excision, and were followed for development of stroke/transient ischemic attack (TIA) [63]. A total of 29 patients had any stroke or TIA. There were 6 intracerebral bleeds, 4 perioperative strokes, 13 ischemic strokes, and six TIAs. The mean onset of postoperative stroke/TIAs was at seven years (incidence of 0.36 percent, 19 events per 5231 patient-years). In all CHADS2 groups, observed ischemic stroke/TIA rates were lower than predicted. Importantly, CHADS2 scores of 2 or greater were associated with increased risk of developing stroke compared with patients with lower scores (hazard ratio [HR] 2.15, 95% CI 0.87-5.32). However, in a separate study of 691 patients who underwent a surgical ablation, CHADS2 did not predict stroke but was related to increased bleeding [64].

●Anticoagulant administration – As for other patients with AF, a DOAC is generally preferred. (See "Atrial fibrillation in adults: Use of oral anticoagulants", section on 'Approach to anticoagulation'.)

Patients who are treated with VKA (eg, those with a mechanical valve) who have a subtherapeutic international normalized ratio after the procedure are treated with early bridging anticoagulation with heparin (intravenous unfractionated heparin or low molecular weight heparin).

Outcomes with surgical left atrial appendage occlusion (with or without a Maze procedure) are discussed further separately. (See "Atrial fibrillation: Left atrial appendage occlusion".)

RECOMMENDATIONS OF OTHERS — The Society of Thoracic Surgeons 2017 clinical practice guideline for the surgical treatment of atrial fibrillation (AF) makes the following strong recommendations [65]:

●Surgical ablation for AF is recommended at the time of concomitant isolated aortic valve replacement, isolated coronary artery bypass graft surgery (CABG), and aortic valve replacement plus CABG to restore sinus rhythm.

●Surgical ablation for AF is recommended at the time of concomitant mitral operations to restore sinus rhythm.

●Stand-alone surgical ablation for symptomatic AF without structural heart disease is reasonable in patients who have failed a class I or III antiarrhythmic medication or catheter-based therapy.

The following recommendations were made in the Heart Rhythm Society/European Heart Rhythm Association/European Cardiac Arrhythmia Society/Asia Pacific Heart Rhythm Society/Latin American Society of Cardiac Stimulation and Electrophysiology (Sociedad Latinoamericana de Estimulación Cardíaca y Electrofisiología) expert consensus statement on catheter and surgical ablation of atrial fibrillation [66]:

●Surgical ablation for AF is recommended at the time of concomitant isolated aortic valve replacement, isolated CABG, and aortic valve replacement plus CABG to restore sinus rhythm if the patient is symptomatic and refractory or intolerant to one class I or III antiarrhythmic medication.

●Surgical ablation for AF is recommended at the time of concomitant mitral operations to restore sinus rhythm for all symptomatic AF patients.

●Stand-alone surgical ablation is reasonable for persistent and long-standing patients who have failed one or more attempts at catheter ablation who prefer a surgical approach after review of safety and efficacy of options. For paroxysmal AF, it may also be considered after one or more catheter attempts.

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.)

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

SUMMARY AND RECOMMENDATIONS

●Several surgical techniques have been developed for the control of refractory atrial fibrillation (AF) and maintenance of sinus rhythm. In most patients, these techniques are employed as adjunctive therapy in patients undergoing other cardiac surgery for some other reason, such as mitral valve or coronary artery bypass surgery. (See 'Indications' above.)

●These surgical procedures appear effective at eliminating or reducing the frequency of AF in a high percentage of patients. (See 'Maze procedure' above and 'Endocardial surgical ablation' above.)

●After surgical ablation, we continue anticoagulation in patients at high risk for stroke.

For all others, we often discontinue anticoagulation two to three months after successful restoration of sinus rhythm and particularly in patients at high risk for bleeding. Patients in persistent AF should receive continued anticoagulation. This practice applies only to patients who have had the left atrial appendage removed or ligated.

●Patients with recurrent AF after one of these procedures may be candidates for electrophysiology study and catheter ablation. (See 'Treatment failures' above and "Atrial fibrillation: Catheter ablation".)

ACKNOWLEDGMENT — The UpToDate editorial staff thank Alan Cheng, MD, who contributed to earlier versions of this topic review.