Atrial fibrillation (AF) is a supraventricular arrhythmia characterized by poorly coordinated atrial activation and associated with reduced atrial mechanical function. Over the long term, it is associated with an increased risk of stroke, heart failure and mortality. AF can occur in the absence of overt structural heart disease in up to 30% of patients; however, hypertension, coronary artery disease (CAD), valvular heart disease, and congestive heart failure (CHF) are common associated conditions. A clinically useful consensus classification recognizes three patterns of AF: paroxysmal (lasting <7 days and self-terminating), persistent (lasting >7 days and requiring electrical or pharmacological cardioversion), or permanent (cardioversion failed or not attempted).
Focal activity, multiple small re-entrant wavelets, and large macro-re-entrant circuits have all been implicated in AF, influenced by changes in ionic currents and autonomic inputs. Atrial fibrillation can be initiated by ectopic beats originating from the pulmonary veins (PVs) or elsewhere in the atria. Experimental work has shown that a high frequency source is capable of maintaining AF. An alternative mechanism implicates the presence of macro-re-entrant loops. or multiple re-entrant wavelets meandering throughout the atria seeking non-refractory tissue, the number of which is related to atrial refractory period, mass, and conduction velocity.
Preablation and Periprocedural Evaluation
Transthoracic echocardiography to assess cardiac structure and function is required in all patients. Preprocedural transesophageal echocardiography is widely employed to exclude the presence of intra-atrial thrombus. Electroanatomic mapping can localize a given electrode position in 3-D space and thereby construct a representation of atrial anatomy while simultaneously gathering information on electrogram morphology, activation timing and voltage amplitude. Magnetic resonance or tomographic cardiac images acquired prior to the ablation procedure give detailed anatomical information that can be useful to plan the procedure. These images can be imported or 'merged' into the electroanatomic platform and integrated with the mapping-acquired anatomy to aid catheter manipulation and deployment of lesions during the ablation procedure. Such techniques are now widely available, can reduce intraprocedural radiation exposure and may improve procedural outcomes.
Techniques for Ablation
The mechanisms of AF are heterogeneous and may vary particularly according to the extent of atrial fibrosis and the chronicity of AF. Paroxysmal AF, especially of short duration, is frequently a purely trigger-dependent phenomenon whereas persistent and permanent AF generally involve more diffuse abnormalities of the atrial substrate. Elimination of specific triggers of AF in an individual patient requires spontaneous firing to be readily identifiable during an ablation. These are often unpredictable and poorly reproducible. However, the PVs are well established as the dominant sources of triggers in paroxysmal AF, and may contribute to its maintenance. Less common sources of focal activity include the superior vena cava, coronary sinus, vein of Marshall, and other atrial sites.
Appropriate techniques for modification of atrial substrate outside PVs is less clear. Early attempts at catheter ablation of nonparoxysmal AF were inspired largely by the success of the Maze surgical procedure. The first catheter-based, linear ablation attempt to recreate the Maze III lesion set was accomplished by Swartz and colleagues in 1994. These procedures are designed to reduce the mass of atrium available for fibrillation and to interrupt larger reentrant circuits that perpetuate AF. However, extensive ablation and prolonged procedural times were required. At present, there are three major techniques for ablation of AF-PV isolation, left atrial linear ablation and ablation of left atrial electrophysiological targets.
Ablation targeting the pulmonary venous-atrial junction is effective in isolating the left atrium from proarrhythmic electrical sources in the PVs. This has been accomplished either relatively close to the tubular portion of the vein, or as a wider encirclement of each pair of venous ostia, encompassing a larger portion of the posterior left atrium. PV or pulmonary antral isolation confirmed by absence or dissociation of venous potentials is easily demonstrated and provides an objective end-point for effective isolation. Following PV isolation (PVI) alone, success rates of 60-85% have been reported in patients with paroxysmal AF, free of antiarrhythmic drugs during follow-up of 6-18 months. Recurrences of arrhythmia following successful PVI are generally related to recovered conducting tissue at the pulmonary venous ostia. However, in 5% of patients, more organized macro-re-entrant atrial tachycardias can also be identified.
Left Atrial Linear Ablation
As the sole ablation technique, PVI appears less successful in maintaining sinus rhythm in patients with nonparoxysmal AF. In these patients, additional linear lesions at the left atrial roof and mitral isthmus are intended to eliminate additional arrhythmogenic substrate and to prevent large atrial re-entrant circuits potentially involved in perpet-uation of AF. Complete linear lesions have been shown to improve outcomes. Even if incomplete, they may be effective merely by incorporating other abnormal arrhythmic substrates within their path. PV isolation plus ablation at the roof and mitral isthmus achieved sinus rhythm in 69% of patients with persistent AF in comparison with only 20% of patients who underwent PVI alone.The incremental benefit of mitral isthmus ablation in addition to PVI was greater for patients with persistent than with paroxysmal AF. Linear ablation in addition to PV isolation was demonstrated effective in preventing atrial tachycardia following circumferential PV ablation for paroxysmal AF.
Left Atrial Electrophysiological Targets
Fractionated potentials are high frequency, cycle length-dependent signals whose precise mechanism is unclear, but may signify local areas of slow or anisotropic conduction, continuously re-entering impulses or temporal overlap of different activation waves. Nademanee et al. demonstrated that ablation solely targeting areas of continuous fractionated electrical activity was effective in treating both chronic and paroxysmal AF. Similar results have been reported by others.
It is not known whether ablation of all such sites is necessary or if it is possible to target specific locations and thereby limit the extent of unnecessary ablation and resultant tissue damage.
Patient Selection and Outcomes
The feasibility of catheter ablation, using a variety of techniques, has been demonstrated for patients with both paroxysmal and nonparoxysmal forms of AF. Recent data also support the extension of ablation as a treatment option populations initially thought unsuitable for the procedure, including the elderly, and those with moderate to severe LV dysfunction and CHF.
Three trials have demonstrated the superiority of catheter ablation either in combination or in direct comparison with antiarrhythmic medication in patients with drug-refractory paroxysmal AF. A history of symptoms despite therapeutic doses of antiarrhythmic medication (Type 1A, 1C, or III according to the Vaughan Williams Classification) is widely accepted as the minimum qualifying criterion for catheter ablation. This strategy is endorsed by the most recent American College of Cardiology (ACC)/American Heart Association (AHA)/ European Society of Cardiology (ESC) guidelines on management of patients with AF where catheter ablation was recommended for consideration following following failure of antiarrhythmic medication for recurrent paroxysmal AF.
Several groups have demonstrated proof of concept in extending catheter ablation to patients with persistent or permanent AF, achieving medium term (1-2 years) success rates of 70-95% with an acceptable procedural risk in their hands.These procedures usually require a combination of the techniques described above. At present, the extensive procedures required, coupled with the shortcomings of current energy delivery systems (non-transmurality of lesions, tissue recovery), represent significant limitations to effective ablation. As the long-term follow-up data are gathered and tools and techniques improved, the role of catheter ablation will become more clearly defined. In the meantime, published guidelines recommend catheter ablation for recurrent persistent AF only after failure of at least one antiarrhythmic medication and significant symptoms despite rate control.
Complications of AF ablation may arise as a result of direct injury to cardiac structures, thermal injury to adjacent extracardiac structures or thromboembolism. A recent worldwide survey of centers performing catheter ablation reported a major complication in 6% of ablation procedures. Cardiac tamponade has been reported in 1-2% of cases. Injury to the phrenic nerve, the right substantially more often than the left, is observed in 0.5% of cases with complete or partial recovery in the majority. Thromboembolic phenomena occur in 1-2% of procedures, and the risk has declined with recognition of the importance of maintaining adequate anticoagulation during the procedure (ACT >300sec). The most significant complication of LA catheter ablation is atrioesophageal fistula formation, with a reported incidence of between 0.05% and 1% and an associated mortality in excess of 50%.
The esophagus in located immediately behind the posterior left atrial wall, and excessive heating from adjacent radiofrequency energy applications may predispose to injury. Potential solutions to this rare but devastating complication that are currently under evalulation include realtime monitoring of esophageal temperature and location, use of alternative energy sources, and improved early detection of esophageal injury. Historically, PV stenosis was a significant complication of AF ablation occurring in 1-5% of patients undergoing ablation within or closely adjacent to the PVs. However, recent changes in technique, including wider area PV isolation, more remote from the venous ostia, have dramatically reduced the risk of this complication.
Follow-up and Post-procedure Care
Freedom from AF, both symptomatic and asymptomatic, at specified intervals following ablation without the use of concomitant antiarrhythmic medication is the ideal clinical end-point. Consensus is needed on what constitutes adequate monitoring and what is the mimimum acceptable AF burden to satisfy this end-point. Although symptomatic improvement despite continuing AF is a valuable end-point from the patient's perspective, absence of symptoms is clearly not reliable proof of the absence of AF, and therefore of little use in determining future stroke risk and the need for continued anticoagulation. The definition of success declines as the frequency and intensity of monitoring increases. A minimal follow-up protocol includes three-monthly Holter, event monitor and electrocardiogram recordings. However, more extensive recording; with an event monitor for one year with daily recordings when asymptomatic and at any time when symptomatic, have also been employed.
Early recurrence of AF within the initial weeks after ablation occurs in up to 50% of patients. In a majority of such patients, the AF subsides, and periprocedural inflammation subsides and lesions mature. For these reasons, antiarrhythmic drugs are often continued for the initial 1–2 months post procedure. Depending on the specific technique employed, and chronicity of AF, 10–30% of patients will require additional ablation procedures. In patients with no or a single risk factor for stroke, and who have had a successful and uncomplicated procedure, anticoagulation usually can be stopped after three months. The risk of stopping anticoagulation in patients with an initially successful procedure, but multiple risk factors, or a history of prior stroke/transient ischemic attack (TIA) is unclear.
While ablation has been demonstrated to be more effective than medical therapy for eliminating recurrent AF in selected groups of patients, impact on long term morbidity and mortality is less certain. There is a need for a large-scale, prospective clinical trials to determine whether ablation has an associated benefit in terms of stroke risk, heart failure risk, and mortality across a broad spectrum of patients, including the elderly, those with chronic AF, LA enlargement, structural heart disease, and heart failure. Such trials will also help define which patients have the most to gain by restoration of sinus rhythm if acceptable efficacy and safety can be demonstrated. Recovery of conduction across previously ablated tissue is responsible for a large majority of arrhythmia recurrences following catheter ablation for AF. Radiofrequency is the dominant energy source used but remains inadequate to ensure lesion continuity and permanence without an unacceptable increase in procedural time and complications. New, effective and safe alternative energy sources are needed.