Print this Article 
Send to a FriendAbstract
Atrial fibrillation (AF) is the most common cardiac rhythm disturbance and its diagnosis carries significant cardiovascular morbidity and mortality. The medical treatment of AF is cumbersome and unsatisfactory, leading to renewed interest in developing alternative treatments for AF. Building upon the pioneering work of Cox and colleagues, recent reported series have demonstrated the feasibility of treating patients undergoing cardiac surgery for other structural heart disease with limited, left-atrial ablation lesion sets using alternative energy sources. This review summarizes recent advances in surgery for AF that will aid in the development of effective, minimally invasive surgical procedures to cure patients of AF.
Introduction
Atrial fibrillation (AF) is characterized by rapid and irregular activation of the atria, leading to loss of normal sinus rhythm. In AF, various regions of the atrial wall pulse 400-600 times per minute and the ventricular rate is determined by the interaction between the atrial activity and the filtering function of the atrioventricular node. AF is the most common cardiac rhythm disturbance, affecting an estimated 2.2 million people within the United States. The incidence of AF increases with age, with a prevalence of 0.5% of people in the fifth decade rising to 10% of people in the eighth decade [1]. AF is associated with a number of predisposing cardiovascular disorders, including coronary artery disease, valvular heart disease, congestive heart failure, and hypertension. However, in up to 31% of cases AF is not associated with an underlying cardiovascular disorder [2].
AF contributes significantly to cardiovascular morbidity and mortality. It is an independent risk factor for death with a relative risk of approximately 1.5 for men and 1.9 for women even after adjustment for associated cardiovascular disorders [3]. Because of the loss of effective atrial contraction, stasis of blood in the atria predisposes affected patients to thromboembolism. Patients with AF have a five-fold increased risk for stroke compared to age-matched controls, and AF is responsible for as many as 15% of all strokes [3, 4].
The long-term medical treatment of AF with antiarrhythmic drug therapy is associated with a failure rate of 50% at one year and up to 84% at two years [4, 5]. In addition, currently available antiarrhythmic agents are not specific for atrial activity and therefore can have profound effects on ventricular electrophysiology. The medical treatment for AF has therefore largely focused on ventricular rate control and management of thromboembolic risk with oral anticoagulants. While warfarin therapy has been shown to have a decisive benefit in reducing thromboembolism in patients with chronic AF, this treatment is cumbersome and exposes patients to significant hemorrhagic risk [6].
Dissatisfaction with medical therapy has spurred efforts to develop a surgical treatment for AF. The pioneering work of Cox and colleagues has demonstrated the feasibility of treating AF surgically by interrupting the atrial pathways for multiple reentry circuits, which are necessary for the maintenance of AF [7-14]. Surgery for the treatment of atrial fibrillation in isolation has not been widely adopted, however, because the procedure remains complex, time-consuming, and requires cardiopulmonary bypass. Ongoing work has focused on developing a potentially less invasive operation by simplifying the pattern of atrial lesions and evaluating alternative energy sources that can create them quickly, without a cut-and-sew technique. The purpose of this review is to provide an update on recent advances that may facilitate the development of effective, minimally invasive surgical procedures for the treatment of atrial fibrillation.
Pathophysiology
In order to evaluate proposed lesion sets for atrial ablation procedures it is important to have an understanding of the underlying pathophysiology of atrial fibrillation. Since the early twentieth century, the dominant conceptual model of AF has been multiple-circuit re-entry. Moe and co-workers initially emphasized the role of re-entrant wavelets in the perpetuation of AF [15, 16]. Allessie and colleagues subsequently advanced the “leading circle” hypothesis in which re-entering impulses are propagated centripetally, tangentially, and centrifugally [17]. Centripetal impulses encounter refractory tissue and die out while tangential impulses are able to establish re-entrant circuits as local conditions, including atrial size and refractory period, allow [1].
An important further advance in our understanding of AF has been the recognition that AF itself alters atrial electrophysiologic properties—a process called “electrical remodeling”—in such a way that the maintenance of the arrhythmia is favored [18]. The approximately ten-fold increase in atrial rate associated with AF causes intracellular calcium loading. As a compensatory response, atrial myocytes downregulate calcium channel activity, which shortens the action potential duration, reduces the refractory period, and promotes the induction and maintenance of AF by multiple-circuit re-entry [1].
Over the past several years, a number of observations have been made that challenge the conventional viewpoint that all AF results from multiple-circuit re-entry. Experimental mapping studies have suggested the importance of a primary local generator, such as a single small re-entry circuit or ectopic focus [19, 20]. This notion has been supported by the finding that left atrial sources of ectopic activity are of particular importance in a subset of patients. In patients with paroxysmal AF, Haissaguerre and colleagues have shown that arrhythmias originate from ectopic foci in the pulmonary veins up to 94 percent of the time [21]. Local ionic differences may lead to shorter refractory periods in left atrial tissue, favoring re-entry [22]. Collectively, these recent findings challenge the long held view that all AF results from multiple-circuit re-entry and support the rationale for developing surgical procedures for the treatment of AF that focus on the left atrium and pulmonary veins.
Lesion Sets
|
The outstanding results of the Cox-Maze III procedure justify its status as the 'gold standard' surgical procedure for AF. Cox and colleagues report an overall success rate of 99% in curing patients of AF [9]. No instances of sinus node damage have been identified. Left and right atrial function have been documented in 93 and 99% of patients, respectively. High rates of freedom from AF have been reported by other investigators performing the Cox-Maze III procedure [23-25]. Despite this high degree of success, the procedure has not gained widespread clinical application due to its perceived complexity. Furthermore, the overwhelming majority of cases have been performed with the use of cardiopulmonary bypass.
|
The results of recent published series of patients undergoing left-sided partial Maze procedures is summarized in Table I [28-36].
These series in general share a number of limitations. These include relatively low numbers of patients, incomplete long-term follow-up, and non-standardized definitions of surgical success and failure. Gillinov and colleagues have advocated a standardized set of definitions to address the latter limitation [37]. The vast majority of patients in these series had structural heart disease requiring additional procedures to be performed at the time of their partial Maze, including mitral valve replacement and repair and coronary artery bypass.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Table 1. |
Despite these limitations, results of these early series have been an important incremental advance in the development of surgery for atrial fibrillation. These investigators have shown the feasibility of adding partial Maze procedures to complex operations for structural heart disease. This has been accomplished with only a modest increase in procedural time and maintenance of low complication rates [38-43]. Mid- and long-term results have been encouraging, with success rates generally in the 70-80 percent range. Pasic and colleagues have reported success rates as high as 92% at six month follow-up [36]. While the results of partial Maze procedures should be examined in the context of the outstanding experience Cox and colleagues have documented with the Cox-Maze III procedure, these smaller studies have shown the potential for partial Maze procedures to effectively treat atrial fibrillation with a simple, short, and less invasive procedure.
Alternative Energy Sources
One of the obstacles to widespread clinical application of the Cox-Maze III procedure is the procedure time required by the cut-and-sew technique used to interrupt atrial re-entry circuits. The development of alternative energy sources that can ablate atrial tissue by topical application has been a requisite advance in the evolution toward faster, less invasive procedures for atrial fibrillation. The ideal energy source would be fast, reliable, yield a full thickness lesion, would not damage surrounding tissues, and would be amenable to off-pump and minimally invasive application. Recently developed alternative energy sources now possess a number of these desirable characteristics.
Radiofrequency
Radiofrequency energy uses an alternating current from 350 kHz to 1 MHz to heat tissue, resulting in thermal injury [37]. There has been a significant experience accumulated using radiofrequency ablation to treat patients via catheter-based approaches [44-47]. This success has led to the development of a number of radiofrequency energy probes that are useful for application to atrial tissue during cardiac surgery [38-41]. These probes can be applied to either endocardial or epicardial heart surfaces to create transmural linear lesions that block atrial conduction. Radiofrequency ablation by epicardial application holds promise for developing off-pump and minimally invasive procedures for AF.
The majority of radiofrequency ablation procedures to date have been performed with unipolar systems in which the patient is grounded by an indifferent skin electrode. In such systems, current flows from the radiofrequency probe to contacted atrial tissue, where thermal energy is released as a result of resistance to conduction. Unipolar systems have a number of limitations related to the unfocused nature of the energy that is delivered. Local temperatures can exceed 100 degrees C, leading to surface charring and potential thromboembolic complications. Heat is conducted to surrounding tissues, raising the risk of damage to surrounding structures. Esophageal perforations have been reported with this technique [34]. Consistent surface application to make uniform lesions is difficult and there is no indication when a transmural lesion has been made. Recently developed bipolar radiofrequency clamps address these limitations and allow creation of more precise and uniform transmural lesions. The early clinical experience with bipolar radiofrequency ablation has demonstrated consistent conduction block with epicardial and off-pump application [48]. Experimental studies have demonstrated its potential in epicardial, off-pump approaches to AF surgery [49].
Cryothermy
Cryoablation is performed with a nitrous oxide cooled probe that when applied to atrial tissue at -60 degrees C for 2 minutes reliably produces transmural lesions that block atrial conduction. An advantage of this technique is that there is no tissue vaporization or charring and the endocardial surface remains smooth following cryoablation. Cox and colleagues were the first to incorporate this modality into surgery for AF and it remains an important component of the Cox-Maze III procedure [8, 10, 11, 13].
Cox and colleagues have subsequently published results for the 'cryo-Maze' procedure in which the lesions of the Cox-Maze III operation are performed solely by cryoablation [8]. Other groups (Sueda, others) have demonstrated the feasibility of left sided partial Maze procedures using cryoablation [28, 50, 51]. Ongoing development of minimally invasive procedures using cryoablation has been somewhat limited by currently available cryoprobe systems, which are rigid [37].
Microwave
Microwave ablation makes use of high-frequency electromagnetic radiation, which upon application to atrial tissue causes oscillation of water molecules, converting electromagnetic energy into kinetic energy and producing heat. This heat causes thermal injury leading to conduction block. However, unlike radiofrequency heating, microwave heating does not cause endocardial surface charring, which may reduce the risk of thromboembolism [52]. Microwave ablation has greater tissue penetration than radiofrequency ablation, increasing the likelihood of a transmural lesion. These advantages have led to an increasing interest in the use of microwave ablation in surgical procedures for atrial fibrillation. Microwave ablation probes are commercially available and have been successfully used in partial Maze operations to treat chronic AF [52, 53]. [View realmedia MicroMazeTM video clip]
Ultrasound
Focused ultrasound (8-10 MHz) can be used to deliver energy to atrial tissue which results in deep heating, coagulation necrosis, and conduction block. The feasibility of this modality for the treatment of atrial fibrillation in humans has been demonstrated in early catheter-based approaches [54, 55]. Systems used in these studies make use of tubular transducers that deliver cylindrical zones of ultrasonic waves to produce circumferential lesions around individual pulmonary veins [56]. This characteristic lesion pattern has potential applications in minimally invasive surgical approaches to AF. Planar ultrasound transducers could be used to produce linear lesions and for epicardial application [57].
Laser
Light energy has been used experimentally to produce linear myocardial lesions with both neodymium:yttrium-aluminum garnet (Nd:YAG) lasers and an infrared coagulator [58-61]. These methods produce well-demarcated transmural photocoagulation necrosis with relatively low peak tissue temperatures and without tissue vaporization. Light energy from a Nd:YAG laser has been delivered radially through flexible optical fibers oriented parallel to myocardium surfaces to produce long linear lesions in a single application [58, 59]. Light energy from the commercially available infrared coagulator is delivered via a light-conducting quartz rod that is applied to the myocardial surface [60, 61]. The efficacy of both methods has been evaluated experimentally with epicardial as well as endocardial application.
Conclusions
Recent advances in surgery for AF have focused on developing faster, simpler, and less invasive procedures with efficacy similar to that achieved with the Cox-Maze III procedure. A number of investigators have published results of left-sided, partial Maze procedures performed using alternative energy sources to make ablative lesions. While the long-term results of these studies have not matched the standard set by Cox and colleagues with their established procedure, these studies have demonstrated the feasibility of performing ablative procedures with limited lesion sets to effectively treat patients with medically refractory AF.
The modest success of these procedures has established that the complete lesion set developed by Cox and colleagues is unnecessary to successfully treat a subset of patients with AF. Preoperative identification of this subset of patients for whom left atrial sources of ectopic activity may be amenable to a simple left-sided partial Maze would potentially improve success rates of these procedures without further advancements in technique. Patients with paroxysmal AF have a high incidence of ectopic foci originating from the left atrium or pulmonary veins and are likely to fall into this subset of patients amenable to surgical treatment. On the other hand, a number of pre-operative clinical criteria, such as enlarged left atrial size, prolonged duration of AF, electrocardiogram voltage criteria, advanced age, and associated coronary artery disease have been associated with Maze procedure failure. Further advancements in our understanding of the pathophysiology of AF as well as improved electrophysiologic screening of individual patient's ectopic foci, for instance with advanced mapping techniques, will lead to better patient selection and surgical cure rates [62-64].
Further refinements in surgical lesion sets may take advantage of lessons learned from catheter-based ablation techniques. Catheter-based ablation of the left atrium is technically demanding, time consuming, and associated with a number of complications [65]. However, the alternative lesion patterns used by these procedures, such as individual curcumferential and segmental pulmonary vein isolation, may serve as an important guide for modification of partial Maze procedure lesion sets [54-56, 65, 66]. Although the Cox-Maze III procedure was designed to be a generalized operative approach that interrupts all potential re-entry circuits, what may evolve is a more tailored approach in which catheter-based mapping and ablation is combined with surgical ablation and left atrial appendage excision to interrupt the individual patient's re-entry circuit or ectopic focus. Left atrial appendage excision remains an important advantage of surgical approaches that should be included in all partial Maze procedures. In fact, left atrial appendage excision has been proposed as a stand alone treatment for chronic atrial fibrillation because it removes the most important thromboembolic source for patients with chronic AF [67-69].
The usefulness of alternative energy sources to facilitate partial Maze procedures is now well documented. Partial Maze procedures can now be added to complex operations for structural heart disease, such as mitral valve repair, with only 10-20 minutes added to cross clamp times [37]. Emerging technologies that can be used for epicardial and beating heart application and ablation of atrial tissues, such as bipolar radiofrequency probes, show great promise as investigators work to develop truly minimally invasive approaches to surgical procedures for AF [70, 71].
As new procedures are developed that effectively treat AF, have low morbidity, and are minimally invasive, they will be increasingly utilized to restore sinus rhythm permanently in patients afflicted with AF. Rigorous evaluation of developing procedures will be important in order to evaluate competing therapies, including medical, catheter-based, and surgical treatments. A standardized analysis and reporting system has been proposed to facilitate the fair comparison of emerging treatments for AF [37].
We are arriving at an exciting point in which our understanding of the underlying mechanisms of AF, our increasing clinical experience with surgical procedures for the treatment of AF, and the development of facilitating technologies are all converging to make the development of efficacious, minimally invasive operations for AF within reach. Innovative application of emerging technologies will allow us to offer new surgical procedures that advance the treatment of AF and benefit our patients by being less invasive and more effective.