HomeAbout Dr. WellsContact

Atrial fibrillation

Atrial fibrillation (AF) is the most common arrhythmia encountered in patients. When the atria fibrillate, they beat at 400-600 times per minute. Because this is so fast, the atria just quiver rather than help fill the ventricles. Fortunately, it is the left ventricle that provides cardiac output to the body. How many of the beats are allowed to conduct through the AV node determines the pulse in AF.

Types of atrial fibrillation

There are three main type of atrial fibrillation. The first type is called paroxysmal because patients with this type will go in and out of AF on their own. The episodes can last seconds, minutes, hours or days. They rarely require a cardioversion since they will go back into normal sinus rhythm on their own. The second type of AF is called persistent. Patients with this type of AF may be out of rhythm for long periods of time even weeks or months. They require medications or cardioversion to get them back into sinus rhythm, since they will not convert on their own. The third type of AF is called permanent in that no matter what medications or how many cardioversions are tried; normal sinus rhythm cannot even be restored.

Consequences of atrial fibrillation

When AF occurs, cardiac efficiency declines. This occurs because of several factors. First, when the atrium is no longer coordinated with the ventricle, the stroke volume declines. Compensatory processes occur to support stroke volume via the Frank-Starling mechanism: elevated left atrial and pulmonary pressure augment ventricular filling. However this tends to produce symptoms of breathlessness at rest or with exertion. Secondly, whenever the pulse is rapid and/or irregular, the efficiency declines further. Lastly, when the atria fibrillate, the walls do not pump as vigorously. After 48 hours, a thrombus can form, usually in the left atrial appendage. If the thrombus dislodges, it has about an 85% chance of being embolized into the carotid arteries causing a stroke. In the other 15% of the time, the clot may embolize into the artery of an organ or into a peripheral artery jeopardizing its viability and requiring emergent vascular surgery or catheter retrieval of the embolus.

Symptoms of atrial fibrillation

Some patients are asymptomatic in AF, having the diagnosis made when a physician finds a fast or irregular pulse. However, most patients have symptoms including palpitation, heart racing, fatigue, shortness of breath and/or chest pain. Light-headedness or syncope is more rare. Sometimes, it is difficult to tell when the AF first started, particularly when the symptoms are not prominent. With other patients, it is fairly easy to tell when AF started.

Origin of atrial fibrillation

Over 90% of atrial fibrillation comes from the left atrium, specifically from the pulmonary veins. These pulmonary veins (PV) are the most common source in patients who have paroxysmal AF. Until recently, we assumed that the left atrial muscle stopped precisely where the PVs meet the left atrium. We now know that threads of atrial tissue called fascicles can go up into the PVs for some distance. These fascicles which may contain pacemaker-like cells (similar to those in the sinus node and AV node) are where AF originates. The fast firing follows the fascicle to the left atrium and causes AF. More rarely other veins may be the source of AF such as the vein or Marshall (a small branch off the coronary sinus vein that runs along the outside of the left atrium), the superior vena cava or the inferior vena cava or even the left atrium itself. Most rarely, AF may originate from the right atrium. In patients with persistent or permanent AF, the AF frequently comes from parts of the left atrium itself rather than from the pulmonary veins. This is important to understand if one wishes to understand the different types of curative ablation procedures for AF.

Risks of atrial fibrillation

Atrial fibrillation, per se, is not life threatening, but there is a risk of stroke. Large studies have suggested that patients with AF do not live quite as long as those without. However, most electrophysiologists believe that this risk of death is because of the risk of stroke in some patients. Therefore, once high-risk patients are treated with anticoagulants (1), there is not a difference in longevity. AF does not cause myocardial infarction, a cardiac arrest or death. Therefore AF is a nuisance – there are millions of people who lead decades of their lives in AF.

Treatment of atrial fibrillation

It should therefore not be surprising that restoring sinus rhythm and preventing AF will not prolong life. Rather, such treatment of AF with antiarrhythmic medications or ablation procedures always has some risk. Antiarrhythmics may rarely cause a cardiac arrest, and ablation procedures are associated with death and other risks. There are only two other reasons to try to restore and maintain sinus rhythm in patients with AF.

The first of these is to improve quality of life. Restoring sinus rhythm will prevent the symptoms associated with AF. However, if the antiarrhythmic medications required have their own side effects, these may in some cases be worse than the symptoms from the AF itself. Nevertheless, many patients feel better once out of AF.

The second reason to restore and maintain sinus rhythm is to prevent the rare occurrence of a dilated cardiomyopathy that can occur if the pulse remains fast (usually averaging 130 beats per minute for at least 90% of the day). However usually this can also be prevented by giving AV nodal blocking medications (digitalis, calcium blockers or beta blockers) to slow the ventricular response to AF.

An important consideration in treating patients with AF is the presence of structural heart disease. Hence an echocardiogram is usually performed to look for valvular heart disease, left ventricular hypertrophy, pericardial effusion that might indicate pericarditis, left ventricular systolic function (ejection fraction) and left atrial size. If the patient has a regional left ventricular wall motion abnormality indicating a myocardial infarction or if there is cardiomyopathy, then some medications like flecainide and propafenone are too risky. The bigger the left atrium, the smaller the chance that normal rhythm can be safely restored and maintained by medications or ablation procedures. For example, if the left atrium is larger than 60mm, only amiodarone has a reasonable chance of working (possibly dofetilide).

Symptomatic patients can be treated with one of two strategies. With the rate control strategy, no effort is made to restore sinus rhythm. Rather digitalis, calcium and/or beta-blockers are given to slow the pulse. If single agents are ineffective then combinations may be additive or even synergistic. The goal is to get resting heart rate into the 60’s and maximal exercise heart rate about 150. If medications are not effective or cause side effects, then one can switch to a rhythm strategy (see below), or the AV node can be ablated and a permanent pacemaker placed.

The other strategy is the rhythm control strategy. With this approach, cardioversion and antiarrhythmic drugs are employed to restore and maintain sinus rhythm, or ablation procedures can be performed to cure AF (see below). When curative left atrial ablation for AF was compared with AV node ablation and a permanent pacemaker in elderly patients in a nonrandomized study (patients chose which therapy they wanted), the patients who received AV node ablation and a pacemaker were less likely to be on antiarrhythmic drugs (3% vs. 30%) but more likely to be in permanent AF (69% vs. 8%) and more likely to have congestive heart failure (53% vs. 24%) (2).

Asymptomatic patients with AF and patients in whom the rhythm control strategy is felt to be unlikely to work or unsafe are usually left in AF and a rate control strategy employed to prevent the development of a rate-related cardiomyopathy. If medications are ineffective in slowing the resting pulse or cause side effects, then the AV node can be ablated (see below) and a permanent pacemaker placed. The cost of treating a patient with AF ranges from $2000 US to over $10,000 US each year (3).

Left atrial ablation for cure

Curative ablation for AF has been the most difficult and risky ablation procedure performed. However, in recent years, the procedure has become more safe and effective such that in 2007, ablation for AF is a viable alternative for many patients. Patients are opting for this procedure sooner than in the past (4). Because the procedure is complex, as much information about the patient’s left atrial size and shape as possible is necessary. Typically a CT angiogram is performed. The three dimensional model of the left atrium can be imported from the CT scan into a special 3D mapping system which will be used during the AF ablation procedure. In this way, the electrophysiologist performing the ablation will have a map of the left atrium and the PVs – this helps them know where to make the ablation lesions.

Preparation for the ablation

If anticoagulants are stopped, this typically occurs a few days before the procedure, although some centers do not stop it at all. Antiarrhythmic drugs might or might not be stopped. A transesophageal echo (TEE) is performed after the patient is anesthetized. Usually there is no left atrial clot, and the ablation procedure ensues. If left atrial clot is present the procedure is cancelled, and the patient receives more intense anticoagulation for a few months and the TEE is then repeated to see whether ablation is now safe.

Types of curative focal RF ablation for AF

Pulmonary Vein Isolation (PVI) is the most commonly performed procedure for AF. In this procedure, IV sheaths in the femoral and/or subclavian/internal jugular veins are used. Catheters are passed to the right atrium and coronary sinus vein. Then a specially shaped long sheath is used which accommodates a needle. A transseptal catheterization is performed by using the needle to puncture the septum between the right and left atrium. Through the needle a guide wire is passed and then the sheath can be pushed into the left atrium. This procedure is usually repeated placing a second sheath. Through one of the sheaths, a special mapping catheter called a halo catheter is positioned right where the PV enters the left atrium. This catheter shows electrical signals indicating where the fascicle(s) enter the left atrium. The RF ablation catheter has a 4-8 mm metal tip electrode capable of making a focal lesion about 6x6 mm in area. RF ablation at this and other left atrial sites where the fascicles are located will destroy them and “isolate” the vein. Once isolated, there can be firing within the vein, but it cannot reach the heart, and the rhythm remains normal. Usually all four PVs are studied. Care must be taken to avoid burning up in the PV or stenosis may occur. A mesh catheter is in testing that is capable of making a circular RF lesion (5).

This procedure is best suited to patients with paroxysmal AF who have a normal-sized left atrium. After one or two procedures, about 70% of patients are cured in experienced centers. When this procedure was compared with antiarrhythmic drugs, 86% of people having PVI were free of atrial fibrillation vs. 22% of those receiving medications (6).

The next-most commonly performed ablation procedure for AF is called wide area circumferential ablation or WACA. In this procedure, multiple burns using RF energy are placed to make a complete circle around all of the four PVs. This also isolates the PVs so that firing cannot reach the left atrium. The risk of PV stenosis is essentially zero in this procedure. The other risks are similar to PVI listed above. However, gaps in the ablation lines can occur and lead to atypical atrial flutters that can be difficult to cure (7). Overall the cure rate of this procedure is nearly 90% in experienced centers.

Focal Left Atrial Ablation or FLAA is a third ablation procedure that is better suited to patients with persistent or permanent AF with larger left atrial size. These patients are seldom cured by PVI or WACA. In FLAA, the sites in the left and right atrium and veins that cause AF are localized by recording very rapid, complex, fractionated electrical signals. These areas are assumed to be the drivers for AF and are burned with RF energy. PVI or lines of burns are sometimes performed as well. In a group of AF patients undergoing FLAA 83% were cured (8) in one series and 77% in another series (18% of patients required a second procedure) (9). New experimental mapping systems will likely make this procedure more effective in the future.

Linear left atrial ablation or LLAA is the last type of ablation and again is better suited for people with persistent/permanent AF. A randomized study compared LLAA with FLAA and found them both to cure 60-68% of patients – they were equally safe (10), although theoretically, LLAA avoids the risk of an atrioesophageal fistula (see below).

Risks of curative focal left atrial RF ablation

The cumulative risk of curative AF ablation is about 6% in low-volume centers and about 3% in high-volume centers - individual risks include the following:

  1. Death
  2. Stroke
  3. Perforation of the heart with tamponade
  4. PV stenosis
  5. Injury to other blood vessels
  6. New arrhythmias such as atypical atrial flutter
  7. Left atrial-esophageal fistula
  8. Injury to the phrenic nerve

Death, fortunately, occurs rarely. In the largest study of 8745 patients worldwide (8) the risk was .05%. Two of the deaths were strokes, 1 a perforation of the heart and 1 unknown.

Stroke occurred in about 1% of curative ablations for AF. The risk can be minimized by not stopping the anticoagulant medications and performing a TEE before ablation to be sure that there is no blood clot in the left atrium. Also using high doses of an intravenous anticoagulant, heparin during the procedure and continuing the anticoagulant pills after the procedure lowers risk. Making sure that intravenous fluids are flowing through the sheaths also lessens the risk of a blood clot with subsequent stroke.

Perforation of the heart occurred in 1.2% of curative AF ablation procedures and led to one death. If the heart is perforated, blood enters the pericardial space and can cause tamponade. Percutaneous catheter pericardiocentesis usually suffices to drain the blood, but in rare cases open-heart surgery is required to over sew the hole.

PV stenosis used to occur when ablation occurred up in the PV. It was then recognized that using lower temperatures and power and ablating within the left atrium (not up in the vein) lowered the risk to about 1-4% (11). Some patients with PV stenosis are asymptomatic while but most have symptoms of shortness of breath, cough, chest pain, or hemoptysis. If PV stenosis does occur, it may be difficult to treat. Just ballooning it results in the narrowing recurring about 73% of the time. Using stents reduces the risk of re-narrowing down to slightly less than 50%. Some unfortunate patients have even required heart-lung transplantation for this problem. This risk is only associated with the PVI procedure because RF ablation occurs close to the vein.

Injury to blood vessels occurs in about 1.5% of cases. Usually these injuries require surgery, but they are rarely fatal.

New arrhythmias occur in 3-27% of curative AF ablations. They are associated with all three types of ablation procedures. The risk is higher with WACA with or without FLAA. These may consist of atrial tachycardias from an un-ablated irritable spot or atrial flutters from gaps in ablation lines. Sometimes they will go away on their own, but in most cases, repeat ablation or antiarrhythmic medications are required. These arrhythmias are usually difficult to treat with medications (7).

A fistula between the left atrium and the esophagus occurs in about .05% of curative AF ablations: 50% are fatal. It is unusual to see this complication unless ablation is performed across the posterior wall of the left atrium, since it is lying right on top of the esophagus. Such unfortunate patients usually present 2-3 days after ablation with chest pain, fever, thromboembolism or fatal hemetemesis. This must be suspected early in patients with these symptoms and requires urgent surgery to keep the patient from dying.

Another complication, injury to the right phrenic nerve, paralyzes the right hemidiaphragm. This occurs in 0.1% of curative ablations for AF that target the right-sided PVs. The right lung is responsible for 55% of the pulmonary function, so this is a significant complication. Sometimes the damage is temporary and other times permanent (12).

Cryoablation for atrial fibrillation

PVI with cryoablation using regular catheters has not been very effective. In one study, 52 patients with either paroxysmal or persistent AF underwent attempted PVI with 94% of procedures successful at the end of the ablation session, but AF recurred in 34 patients who had their rhythm medications stopped (13). Complications occurred in 4 patients, but no patient had evidence of PV stenosis over the 12-month follow-up period. A second study using a special curved catheter for PVI was undertaken in 18 patients (14). This novel catheter allowed isolation of 69% of the targeted PVs with focal cryoablation isolating another 22% of veins. 2 patients suffered complications, but only 22% of patients had no recurrence of AF off antiarrhythmic medications. However, no PV stenosis occurred. In fact even when cryoablation is deliberately delivered within the PV, stenosis usually does not occur (15). The theoretical advantages of cryoablation are:

  1. Less disruption of the endocardium with less blood clot formation,
  2. Less weakening of the heart muscle wall so less risk of perforation,
  3. Low risk of PV stenosis.

Eliminating these risks could theoretically lower the 6% risk of curative RF ablation for AF down to 2.8% for cryoablation in low-volume centers and lower the rate further even in high-volume centers. A cryoballoon catheter is now FDA approved as a result of the STOP-AF Trial. The results with this device are promising (16).

Other ablation modalities

High-frequency ultrasound can be delivered through a balloon catheter to try to isolate pulmonary veins. In one study, 13 of 27 patients had no more AF over 12 months off their antiarrhythmic drugs 17 indicating only modest efficacy. In another study, 2 of 15 patients suffered damage to the right phrenic nerve that did not recover over the follow-up period 18- worrisome for safety. A laser balloon device is also in early animal testing.

After curative left atrial ablation

Sometimes blood thinners are not stopped for ablation – this greatly reduces the risk of stroke. If blood thinners are stopped for the ablation, the patient is started back on coumadin/warfarin or one of the newer blood thinners, and this therapy is continued for a few months to see whether the Afib recurs or not. After this time period, many electrophysiologists feel that anticoagulants can be stopped safely. Preliminary retrospective evidence suggests that it may be safe to stop anticoagulants after a successful left atrial ablation procedure 19 With successful ablation, the left atrial appendage gets smaller hence less hospitable to a blood clot 20 However definitive prospective studies are on-going to assess the safety of this approach ( this strategy has not been shown to be safe when rhythm control is obtained with antiarrhythmic drugs, perhaps because asymptomatic AF still occurs). In some patients with AF, antiarrhythmic medications are continued for weeks to months after the procedure. Others may have their medications stopped. Recurrence of AF in the first few months of the procedure does not necessarily mean that the ablation did not work. Sometimes more time is required to see whether the ablation was successful or not. In one study patients who underwent a successful ablation procedure for AF had improved quality of life 21). In another study those with depressed ejection fractions showed improvement in LV function (22).

Review article

McGuire M. Atrial fibrillation: much heat, a little light. J Cardiovasc Electrophysiol. 2006; 17:1286-1287
A provocative editorial on AF ablation

Natale A, Raviele A, Arentz T, et.al. Venice chart international consensus document on atrial fibrillation ablation. J Cardiovasc Electrophysiol 2007; 18:560-580.
A thorough opinion from the experts in AF ablation

Morady, F. Patient-specific ablation strategy for atrial fibrillation: promises and difficulties. Heart Rhythm 2007;4:1094-1096.
A current opinion on the applications/limitations of AF ablation.

Kottkamp H, and Hindricks G. Complex fractionated atrial electrograms in atrial fibrillation: a promising target for ablation, but why, when, and how? Heart Rhythm 2007; 4: 1021-1023.
The promise of AF ablation for patients with more difficult cases of AF.


  1. Connolly S, Eikelboom J, O’Donnell M et.al. Challenges of establishing new antithrombotic therapies in atrial fibrillation. Circulation 2007; 116:449-455.
  2. Hsieh M, Tai C, Lee S etal. Catheter ablation of atrial fibrillation versus atrioventricular junction ablation plus pacing therapy for elderly patients with medically-refractory paroxysmal atrial fibrillation. J Cardiovasc Electrophysiol 2005; 16: 457-461
  3. Reynolds M, Essebag V, Zimetbaum P et.al. Healthcare resource utilization and costs associated with recurrent episodes of atrial fibrillation: The FRACTAL registry. J Cardiovasc Electrophysiol 2007; 18:628-633.
  4. Gerstenfeld E, Callasn D, Dixit S et.al. Characteristics of patients undergoing atrial fibrillation ablation: trends over a seven-year period 1999-2005. J Cardiovasc Electrophysiol 2007; 18:23-28.
  5. Arruda M, He D, Friedman P et.al. A novel mesh electrode catheter for mapping and radiofrequency ablation delivery at the left atrial-pulmonary vein junction. J Cardiovasc Electrophysiol 2007; 18:206-211.
  6. Pappone C, Augello G, Sala S etal. A randomized trial of circumferential pulmonary vein ablation versus antiarrhythmic drug therapy in paroxysmal atrial fibrillation. J Am Coll Cardiol 2006; 48: 2340-2347
  7. Gestenfeld E and Marshlinski F. Mapping and ablation of left atrial tachycardias occurring after atrial fibrillation ablation. Heart Rhythm 2007; 4:S65-S72.
  8. Cappato R, Calkins H, Shen S, etal. Worldwide survey on the methods, efficacy and safety of catheter ablation for human atrial fibrillation. Circulation 2005; 111: 1100-1105
  9. Oral H, Chugh A, Good E etal. A tailored approach to catheter ablation of paroxysmal atrial fibrillation. Circulation 2006; 113: 1824-1831
  10. Oral H, Chugh A, Good E etal. Randomized comparison of encircling and nonencircling left atrial ablation for chronic atrial fibrillation. Heart Rhythm 2005; 2: 1165-1172
  11. Dong J, Vasamreddy C, Jayam V et.al. Incidence and predictors of pulmonary vein stenosis following catheter ablation of atrial fibrillation using the anatomic pulmonary vein ablation approach: results from paired magnetic resonance imaging. J Cardiovasc Electrophysiol 2005; 16:845-852.
  12. Bai R, Patel D, DiBase L et.al. Phrenic nerve injury after catheter ablation: should we worry about this complications? J Cardiovasc Electrophysiol 2006; 17:944-948.
  13. Tse H, Reek S, Timmermans C, etal. Pulmonary vein isolation using transvenous catheter cryoablation for treatment of atrial fibrillation without risk of pulmonary vein stenosis. Journal of the American College of Cardiology. 2003; 42: 752-8
  14. Skanes A, Jensen S, Papp R, etal. Isolation of pulmonary veins using a transvenous curvilinear cryoablation catheter: feasibility, initial experience and analysis of recurrences. Journal of Cardiovascular Electrophysiology. 2005; 16: 1304-1308
  15. Kenigsberg D, Wood M, Alaeddini J, et.al. Cryoablation inside the pulmonary vein after failure of radiofrequency antral isolation. Heart Rhythm 2007;4:992-996.
  16. Andrade J, Khairy P, Guerra P et. al. Efficacy and Safety of cryoballoon ablation for atrial fibrillation: A systematic review of published studies. Heart Rhythm 2011;8: 1444-1451
  17. Nakagawa H, Antz M, Wong T et.al. Initial experience using a forward directed, high-intensity focused ultrasound balloon catheter for pulmonary vein antrum isolation in patients with atrial fibrillation. J Cardiovasc Electrophysiol 2007; 18:136-144.
  18. Schmidt B, Antz M, Ernst S et.al. Pulmonary vein isolation by high-intensity focused ultrasound: first-in-man study with a steerable balloon catheter. Heart Rhythm 2007; 4:575-584.
  19. Oral H, Chugh A, Ozaydin M etal. Risk of thromboembolic events after percutaneous left atrial radiofrequency ablation of atrial fibrillation. Circulation. 2006; 114: 759-765
  20. Chang S, Tsao H, Wu M et.al. Morphologic changes of the left atrial appendage after catheter ablation of atrial fibrillation. J Cardiovasc Electrophysiol 2007; 18:47-52.
  21. Weerasooriya R, Jais P, Hocini M et.al. Effect of catheter ablation on quality of life of patients with paroxysmal atrial fibrillation. Heart Rhythm 2005; 2:619-623,
  22. Gentlesk P, Sower W, Gerstenfeld E et.al. Reversal of left ventricular dysfunction following ablation of atrial fibrillation. F Cardiovasc Electrophysiol 2007; 18:9-14.