HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Correction (v131,p772) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Patrick M. McCarthy Eugene H. Blackstone Gosta Pettersson Joseph F. Sabik, III Lars G. Svensson Delos M. Cosgrove Kathleen M. Hill Gonzalo V. Gonzalez-Stawinski Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gillinov, A. M. Articles by Natale, A. Search for Related Content PubMed PubMed Citation Articles by Gillinov, A. M. Articles by Natale, A. Related Collections Cardiac - other Electrophysiology - arrhythmias

J Thorac Cardiovasc Surg 2005;129:1322-1329
© 2005 The American Association for Thoracic Surgery

Surgery for Acquired Cardiovascular Disease

Surgical ablation of atrial fibrillation with bipolar radiofrequency as the primary modality

^a Center for Atrial Fibrillation and the Department of Thoracic and Cardiovascular Surgery, The Cleveland Clinic Foundation, Cleveland, Ohio.
^b Department of Qualitative Health Sciences, The Cleveland Clinic Foundation, Cleveland, Ohio.
^c Department of Cardiovascular Medicine, The Cleveland Clinic Foundation, Cleveland, Ohio.

Read at the Thirtieth Annual Meeting of The Western Thoracic Surgical Association, Maui, Hawaii, June 23–26, 2004.

Received for publication June 23, 2004; revisions received October 21, 2004; accepted for publication December 15, 2004.

^_* Address for reprints: A. Marc Gillinov, MD, Department of Thoracic and Cardiovascular Surgery, The Cleveland Clinic Foundation/F24, 9500 Euclid Ave, Cleveland, OH 44195. (Email: gillinom{at}ccf.org).

	Abstract

Top Abstract Materials and Methods Results Discussion Discussion Electronic Appendix 1 Electronic Appendix 2 Electronic Appendix 3 Electronic Appendix 4 Electronic Appendix 5 Electronic Appendix 6 References

 
OBJECTIVES: In studying cardiac surgical patients undergoing atrial fibrillation ablation with bipolar radiofrequency, we sought to (1) quantify the time-related prevalence of atrial fibrillation postoperatively and identify its risk factors and (2) determine time-related ablation failure and its risk factors.

METHODS: From November 2001 to January 2004, 513 patients underwent atrial fibrillation ablation (bipolar radiofrequency alone or with cryothermy) and other cardiac operations. Rhythm documented on 3495 postoperative electrocardiograms was used to estimate the prevalence of and risk factors for atrial fibrillation across time. Ablation failure was defined as occurrence of atrial fibrillation any time beyond 6 months after operation.

RESULTS: Prevalence of postoperative atrial fibrillation peaked at about 1 month, decreased to 13% at 6 months, and gradually increased thereafter. Risk factors associated with increased prevalence varied by time period and included older age (P = .004) for early occurrence, lesion set in permanent atrial fibrillation (P = .02) for late occurrence, and larger left atrial diameter (P = .02) and permanent atrial fibrillation (P < .0001) for occurrence across the entire time span. Freedom from ablation failure was 72% at 12 months. Risk factors for ablation failure included lesion set in permanent atrial fibrillation (P = .001), longer duration of atrial fibrillation (P = .01), and larger left atrial diameter (P = .03).

CONCLUSIONS: Bipolar radiofrequency enables extension of ablation to most cardiac surgical patients with atrial fibrillation. Recurrence is influenced by the type and duration of atrial fibrillation, choice of lesion set in permanent atrial fibrillation, and left atrial size. Early operation, careful choice of lesion set, and left atrial reduction might enhance results.

Recently, there has been great interest in direct surgical ablation of AF, fueled by technologic advances and demonstration that the pulmonary veins and left atrium are the primary sites of drivers of AF.^6–12 Ablation technologies, such as bipolar radiofrequency (RF), enable surgeons to isolate the pulmonary veins and create linear left atrial lesions rapidly and safely.^6–11 However, results of these new procedures require scrutiny. The objectives of this study of cardiac surgical patients undergoing AF ablation with bipolar RF were to (1) quantify the time-related prevalence of AF after operation and identify its risk factors and (2) determine time-related ablation failure and its risk factors.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion Discussion Electronic Appendix 1 Electronic Appendix 2 Electronic Appendix 3 Electronic Appendix 4 Electronic Appendix 5 Electronic Appendix 6 References

 
Patients
From November 2001 to January 2004, 513 patients underwent AF ablation with bipolar RF at The Cleveland Clinic Foundation. This represents 25% of the 2079 patients with preoperative AF undergoing cardiac surgery during this time frame and 67% of the 769 patients who had some form of AF ablation. Patients were identified, and preoperative, operative, and postoperative variables were retrieved from the Cardiovascular Information Registry, which has been approved for research by the institutional review board. Mean patient age was 67 ± 12 years (Table 1). The most common indication for operation was mitral valve disease (69%).

AF Ablation
Bipolar RF
All patients underwent AF ablation with the Atricure (West Chester, Ohio) bipolar RF system. In this system, electrodes reside in the jaws of an atraumatic clamp. Energy delivery is continued until tissue conductance between electrodes decreases and reaches a steady state for 2 seconds, indicating transmural conduction block.^8,9

Lesion sets
Ablation was generally performed after cardiac arrest but before the principal operative procedure. Lesion sets varied over time and were performed at the discretion of the surgeon in a nonrandomized fashion (see Electronic Appendix Figure E1). Four basic lesion sets were used most commonly, and these were termed lesion sets 1, 2, 3, and 4 (Figure 1). In all patients, pulmonary veins were isolated with the bipolar RF clamp.

View larger version (15K): [in this window] [in a new window]   Electronic Appendix Figure E1 Use of various lesion sets (1–4) across the study period.

View larger version (74K): [in this window] [in a new window]   Figure 1. Schematic representation of left atrial lesion sets created with bipolar RF. White ovals represent mitral valve, sets of 4 black ovals represent pulmonary veins, and dashed lines represent sites of ablation. All ablation lesions are created with bipolar RF, except in lesion sets 3 and 4, in which the lesion connecting the left pulmonary vein lesion to the mitral valve annulus is created with cryothermy.

There were differences between patients receiving different lesion sets. Compared with those receiving lesion set 1, patients receiving lesion sets 2, 3, and 4 were more likely to have a mitral valve procedure; in addition, those receiving lesion sets 3 and 4 were less likely to have paroxysmal AF. Compared with patients receiving lesion set 1, patients with permanent AF receiving lesion sets 2, 3, and 4 were more likely to have mitral valve disease and left atrial enlargement and less likely to have coronary artery disease (see Electronic Appendixes 1–4).

Follow-up
Postoperative electrocardiograms (ECGs) were used to assess AF. ECGs were performed routinely before discharge and at the discretion of referring physicians during follow-up. Recommended ECG follow-up intervals were 1 month, 3 months, 6 months, 1 year, and annually thereafter. A total of 3495 postoperative ECGs were retrieved. Every patient had at least one record available for analysis (range, 1 day to 25 months). One hundred ninety-five patients had an ECG 6 months or more after operation (Figure 2 and see Electronic Appendix 5, available at www.mosby.com/jtcvs). For purposes of analysis, any one of the following 3 rhythms was considered as postoperative AF: AF, atrial flutter, or a paced rhythm with underlying AF or atrial flutter.

View larger version (21K): [in this window] [in a new window]   Figure 2. Number of patients with ECG follow-up available at and beyond various time points and number of ECGs available for analysis at and beyond these same time points.

Analysis and Depiction of Data
Prevalence of and risk factors for AF
Because there is no practical method for continuous heart rhythm assessment, a novel statistical technique was developed to depict the prevalence of AF versus time on the basis of discrete ECG recordings. We attempted to solve multiple challenges simultaneously: repeated ECG recordings for each patient (repeated measures), variable times of recording, censoring by death, and a complex temporal pattern of AF prevalence that likely represented the variable effect of different modulating factors across time. The analytic approach taken was temporal decomposition, using several simple additive components (early, constant, and late in this instance) into which separate streams of modulating variables were incorporated for simultaneous analysis. The mathematic models used were derived from those previously used over the last 20 years for decomposition of time-to-event data.¹⁴ Longitudinal binary logistic regression analysis for repeated measurements (PROC NLMIXED, SAS, Inc, Cary, NC) was used to resolve the number of time components and to estimate the shaping parameters of each.

Multivariable analysis was performed in the odds domain to identify risk factors for each component (see Electronic Appendix 6, available at www.mosby.com/jtcvs). Because of the limited ability of PROC NLMIXED to explore multivariable relations, we screened variables using ordinary multivariable logistic regression analysis (PROC LOGISTIC, SAS) with a liberal retention criterion (P .15). Candidate risk factors identified were entered at once into PROC NLMIXED and then eliminated one by one until all variables remaining had a P value of .1 or less.

Ablation failure and risk factors
Ablation failure was defined as any occurrence of AF or atrial flutter 6 months or more after ablation. Therefore, to be at risk of ablation failure, patients had to have an ECG obtained in this timeframe. Patients were considered to have ablation failure even if late AF was transient. Data were considered interval censored, with the event assumed to occur between the time of an ECG showing AF and the immediately preceding ECG not showing AF.

To identify risk factors for ablation failure,¹⁴ a parametric model was used to resolve a number of instantaneous risks of the interval-censored event (hazard function) and to estimate shaping parameters. (For additional details, see http://www.clevelandclinic.org/heartcenter/hazard.) Variable selection was by bootstrap bagging^15,16 using a retention criterion of P less than .05 and considering variables appearing in 50% or more models as reliably significant at this level.

Presentation
Predicted percentages of prevalence of AF and freedom-from-event curves are accompanied by asymmetric 68% confidence limits (CLs) comparable to ±1 SE.

	Results

Top Abstract Materials and Methods Results Discussion Discussion Electronic Appendix 1 Electronic Appendix 2 Electronic Appendix 3 Electronic Appendix 4 Electronic Appendix 5 Electronic Appendix 6 References

 
In Hospital
There were no ablation device-related complications. Hospital morbidity included stroke in 8 (2%) patients, transient ischemic attack in 4 (1%), and reoperation for bleeding in 25 (5%). Strokes were judged to be related to the patient’s underlying disease (atherosclerosis or calcified valvular heart disease requiring extensive debridement). There were 10 hospital deaths (2%; CL, 1.3%-2.8%).

Heart Rhythm
Prevalence of AF and its risk factors
A curve demonstrating the time-related predicted prevalence of AF or atrial flutter after operation is depicted in Figure 3. Early prevalence of AF peaked at 38% (CL, 35%-41%) at 2 weeks. By 6 months, the prevalence decreased to 13% (CL, 12%-15%), gradually increasing thereafter to 16% (CL, 13%-19%) at 1 year. At 6 months, 26% of patients were taking antiarrhythmic medications (see Figure E2), and 50% were taking warfarin (see Figure E3).

View larger version (12K): [in this window] [in a new window]   Figure 3. Temporal pattern of AF after ablation on the basis of postoperative ECGs. The solid line represents point estimates enclosed within 68% bootstrap CLs.

View larger version (11K): [in this window] [in a new window]   Electronic Appendix Figure E2 Prevalence of class I and III antiarrhythmic agent use after ablation. The horizontal axis begins at 3 months, marking the first systematic follow-up of patients. Filled circles represent observed prevalences, but only approximate the data because of multiple follow-up inquiries per patient. The solid curve represents the prevalence estimate from the logistic mixed model, and dashed lines represent 68% confidence limits.

View larger version (10K): [in this window] [in a new window]   Electronic Appendix Figure E3 Prevalence of warfarin use after ablation. Format is as for Electronic Appendix Figure E2.

View larger version (10K): [in this window] [in a new window]   Electronic Appendix Figure E4 Decomposition of temporal pattern of atrial fibrillation (AF) after ablation, based on postoperative electrocardiograms, demonstrating 3 phases of prevalence: an early peaking phase, a constant phase, and a late rising phase.

View larger version (15K): [in this window] [in a new window]   Figure 4. Nomograms from the multivariable equation (Table 2) demonstrating effects of factors on prevalence of AF. A, Type of AF. B, Lesion set in paroxysmal AF. There is no difference between lesions sets. C, Lesion set in permanent AF. Note that 1+ refers to both lesion set 1 and miscellaneous additional lesions, generally to the left atrial appendage. Results with lesion set 2 are inferior.

View larger version (13K): [in this window] [in a new window]   Electronic Appendix Figure E5 Nomogram from the multivariable equation (Table 2) demonstrating the effect of 2 left atrial diameters.

View larger version (12K): [in this window] [in a new window]   Figure 5. Freedom from ablation failure. The solid line represents parametric estimate enclosed within dashed 68% CLs. Numbers in parentheses represent patients remaining at risk. The horizontal scale begins at 6 months postoperatively because patients are not at risk for ablation failure until that time.

View larger version (23K): [in this window] [in a new window]   Electronic Appendix Figure E6 Effect of (1) lesion set, (2) type of atrial fibrillation (AF), and (3) duration of AF (along horizontal axis) on estimated 1-year freedom from ablation failure (vertical axis) based on multivariable analysis (see Table 3). Dashed lines enclose 68% confidence limits.

	Discussion

Top Abstract Materials and Methods Results Discussion Discussion Electronic Appendix 1 Electronic Appendix 2 Electronic Appendix 3 Electronic Appendix 4 Electronic Appendix 5 Electronic Appendix 6 References

AF in Cardiac Surgical Patients
The rationale for restoring sinus rhythm in patients presenting for cardiac surgery includes (1) improving survival, (2) reducing risk of stroke and other systemic embolism, (3) eliminating need for anticoagulation, (4) reducing symptoms associated with tachycardia, and (5) improving cardiac output.^1–3 Because most of these patients do not revert to sinus rhythm with treatment of their primary cardiac lesion (with the possible exception of AF of short duration), we took an aggressive approach to ablation.¹⁷

Although the Cox maze III procedure restores sinus rhythm or an atrioventricular paced rhythm in 75% to 95% of patients undergoing concomitant cardiac surgery,^1–5 it has not been widely performed. Recently, there has been great interest in treating AF surgically, which is attributable to increased understanding of its pathogenesis and development of new ablation technologies. In most patients, triggers and substrates for AF are located in the left atrium and pulmonary veins,¹² and therefore ablation is now focused on these areas. Alternative energy sources enable rapid creation of lines of conduction block without the laborious cutting and sewing of the Cox maze III procedure. Reported ablation success ranges from 60% to 95% using a variety of different energy sources and lesion sets in various types of patients.^6,7,10,11 However, there are few data identifying patient- and procedure-related risk factors for ablation failure, and no valid statistical technique to assess results.⁷

Prevalence of AF, Ablation Failure, and Risk Factors
In the current study, bipolar RF was the predominant modality used to create left atrial lesions; cryothermy was used to create lesions on the right and left atrial isthmuses in some patients. Four basic left atrial lesion sets were used. This variability in lesion sets, coupled with varying patient characteristics and types and durations of AF, facilitated identification of risk factors for return of AF and ablation failure.

Risk factors for AF prevalence and ablation failure were both patient and procedure related. Longer duration of AF, permanent AF, and larger left atrial diameter increased either the prevalence of AF, the risk of ablation failure, or both. In fact, these characteristics tended to occur together in the same patients. Others have demonstrated increased risk of ablation failure when these factors were present.^18–22 These findings suggest that it is unlikely that "one size fits all" in ablating AF²³; rather, a tailored approach is indicated.

Procedural risk factors were related to choice of lesion set. In patients with paroxysmal and persistent AF, all lesion sets produced equivalent results. Each lesion set included pulmonary vein isolation, suggesting that drivers of AF in these patients might reside in the pulmonary veins. In contrast, choice of lesion set affected results in patients with permanent AF. Those who received lesion sets 2, 3, and 4 tended to have mitral valve disease and left atrial enlargement (see Electronic Appendix 4 available at www.mosby.com/jtcvs). Lesion set 2, which does not include a lesion to the mitral annulus, produced inferior results in these patients compared with lesion sets 3 and 4. It has been suggested that failure to ablate the left atrial isthmus jeopardizes success.^24,25 Our results support this concept in patients with mitral valve disease. As in previous reports, addition of right atrial lesions (eg, isthmus lesions) did not affect results.²⁶

Others have reported that lesion set affects results. Gaynor and colleagues²⁷ found that the Cox maze III procedure was more effective than the Cox maze I and II. Data from catheter-based procedures in patients with paroxysmal AF suggest that the more left atrial area isolated, the better the freedom from AF.^28–30

Limitations
This series contains early follow-up of a large cohort of patients undergoing intraoperative RF ablation with or without cryothermy. ECGs used for analysis were obtained routinely in the hospital but opportunistically after hospital discharge. Although only 195 patients had ECGs beyond 6 months after operation, there are a sufficient number for analysis. Patients having late ECGs might be selected because they had symptomatic AF, leading to overestimation of the prevalence of AF and ablation failure. Alternatively, asymptomatic AF episodes without ECG verification might have been missed, leading to underestimation of events.³¹ Although periodic Holter monitoring and event recorders would have provided more data for analysis, these methodologies might still fail to capture all AF episodes. Limitations in the ability to document heart rhythm continuously over time led us to develop novel statistical methods that we believe provide accurate estimates of AF prevalence. Although not perfect, this method represents an important improvement over analyses relying on rhythm at last follow-up or patient self-reporting.

Results of AF ablation were assessed and presented in 2 related but different fashions. We estimated the prevalence of AF to depict the predicted prevalence of AF in the population at any given time after ablation. In such a depiction, a given patient might have AF on one ECG and sinus rhythm on the next, unlike in a time-to-event analysis. We also estimated freedom from ablation failure, with ablation failure defined as any recurrence of AF beyond 6 months after operation. Although this definition is stringent, it is appropriate.

The analyses of prevalence of AF and ablation failure do not take into account antiarrhythmic medications that patients might have been taking. The prevalence of antiarrhythmic medication use is depicted separately. Although we recommend discontinuing antiarrhythmic agents 3 months after ablation, continued use is at the discretion of referring cardiologists, which hampered our ability to report the prevalence of AF and freedom from ablation failure off medications.

We did not assess lesion integrity using pacing or other methods; it is therefore possible that operator error or other factors might have resulted in some patients receiving incomplete lesions, which might have contributed to ablation failures.

Clinical Inferences
AF is common early but is transient after intraoperative ablation. Early operation after the development of AF and left atrial size reduction in those with enlarged left atria might improve results. In patients with paroxysmal and persistent AF, simple pulmonary vein isolation might be adequate. In patients with permanent AF, we recommend the left atrial lesion set of the Cox maze III procedure. A prospective randomized clinical trial with longer follow-up is necessary to test these strategies.

	Discussion

Top Abstract Materials and Methods Results Discussion Discussion Electronic Appendix 1 Electronic Appendix 2 Electronic Appendix 3 Electronic Appendix 4 Electronic Appendix 5 Electronic Appendix 6 References

 
Dr Donald B. Doty (Salt Lake City, Utah). Dr Gillinov and his associates have provided us with information on a large series of patients, 513, having ablation procedures with bipolar RF for the treatment of AF during slightly more than a 2-year period. It is a retrospective study and is subject to all of the problems associated with such studies. One, there was no protocol or agreement as to the lesion pattern in these patients. Choice was by surgeon discretion, and the lesion pattern changed and evolved over time. The patients were classified according to established American Heart Association and American College of Cardiology criteria: paroxysmal, meaning easily or spontaneously converted to normal sinus rhythm; persistent, probably can be converted; and permanent, cannot be converted. But the lesion pattern choice was apparently not consistently applied by category of patient.

The title of the article implies that bipolar RF energy was the energy used to create left atrial ablation lesions, but the data show that patients having lesion patterns 3 and 4 had the left atrial isthmus lesion created by means of cryothermia and that one third of the patients had right atrial inferior isthmus ablation lesions created by means of cryothermia, presumably to prevent postoperative atrial flutter.

Data have been analyzed by using complex statistical methods in an attempt to compensate for events occurring over time and in which they used analysis of ECGs obtained nearly randomly in the postoperative period. Terms such as longitudinal binary logistic regression analysis and bootstrap bagging are used. Now Gene Blackstone has explained this to me on frequent occasions, making nearly perfect logical sense, and I accept these methods knowing Gene Blackstone is correct, although I confess I really do not understand them very well, and I am not even going to ask Marc to try to explain them to us.

Dr Gillinov. But I am ready to explain them.

Dr Doty. Well, you do not have to. We accept them. And I think the authors have done a very good job of making some sense out of some difficult data, nearly 3500 ECGs. Let me give you what my interpretation of these data is based on reading the article and hearing what is presented, which quite accurately matches the article. Except for the connecting lesion, lesion patterns 1 and 2 are the same. These are patients who had in common pulmonary vein isolation lesions, and they worked well with the paroxysmal or persistent types of AF, in which one would expect trigger points in the pulmonary veins to initiate the arrhythmia. Lesion pattern 2 added the connecting lesion, but this was not enough when it came to permanent forms, and it has biased the data because the patients who had lesion pattern 2 were heavily weighed with patients with mitral valve disease, whereas the patients with lesion pattern 1 were not, and therefore one would expect the data to come out that way because the substrate for patients receiving lesion pattern 2 implies a much more complex pathogenesis of the AF than trigger points in the left pulmonary veins.

Now lesion patterns 3 and 4 are also the same, having in common pulmonary vein isolation and left atrial isthmus lesions, except that lesion pattern 4 has a double connecting lesion, which, by the way, leads to a portion of the left atrial posterior wall isolated from any electrical depolarization. And this appears to be a more effective lesion pattern when the left atrium is abnormal, such as in mitral valve disease. We do not know from the data, and we have not heard anything presented, what the effect of right atrial isthmus lesions, which were done by means of cryothermia in about one third of these patients, might be, and I would say that the conclusion of the authors, proposed effect of left atrial reduction, is purely speculative because there are no data in this presentation or the article on which you can base that conclusion.

The first question is as follows: What is the added cost of using 2 energy sources for this operation? Second, what is the role of the right atrial isthmus ablation in the treatment of AF? Three, should pulmonary vein isolation lesions be restricted to patients having paroxysmal forms or perhaps persistent forms of AF? And finally, what is the currently recommended lesion pattern for permanent forms of AF?

Dr Gillinov. Thank you, Dr Doty, for all of your comments. I think the preamble to your questions hits on something that can be used as the strength of this presentation. It is incredibly heterogeneous. There were 10 surgeons handling many different types of patients with different lesion sets in different approaches. At first glance that might produce just chaos, but on the other hand, if you have enough patients and enough follow-up, it allows you to discern patterns that relate to both the sort of patients and the lesions created, and you can get the beginnings of some answers, and therefore the heterogeneity was useful.

Regarding the cost of the 2 energy sources? Well, it is higher than we would like, but we used the older reusable cryo system, so that is a sunk cost from 15 years ago. It does not cost us anything additional to use the Cooper Surgical or Frigitronics system once we have already purchased it, and therefore we are paying for only one disposable ablation unit per case. There are people who create the isthmus lesion with RF. We have not because of fear of injury to the circumflex artery, although I am not sure that has been reported.

What about the right atrial isthmus lesion? We analyzed that to determine whether its presence or absence is a risk factor for failure. In our experience the answer is that no, it has no effect, and therefore we create one now only in someone who has a history of flutter.

Should pulmonary vein isolation be restricted to patients with paroxysmal AF? Not entirely. Pulmonary vein isolation is now our lesion set of choice with excision of the appendage in the paroxysmal patient. However, in patients who are very ill and who undergo a complex operation in which we do not want to spend even the additional 10 minutes to open the atrium and create connecting lesions, we will just do the pulmonary veins, and about 40% or 50% of those might have some success. On the other hand, if a patient has permanent AF and there is no specific contraindication or difficulty to opening the atrium and making connecting lesions, then our lesion set of choice in the left atrium looks like a maze procedure.

Dr Vaughn Starnes (Los Angeles, Calif). It is not only a very heterogeneous group of surgeons but also a heterogeneous population of patients.

My question pertains to the mitral valve repair itself. Did the repair and durability of that repair affect in any way the results of your AF; that is, if you had some residual mitral regurgitation or regurgitation developed, did it affect your outcome?

Dr Gillinov. I do not know whether recurrent mitral regurgitation influences the results; it very well might. Because recurrent mitral regurgitation and return of AF are both time-related outcomes of the surgical procedure, analysis of their relationship is statistically challenging.

	Electronic Appendix 1

Top Abstract Materials and Methods Results Discussion Discussion Electronic Appendix 1 Electronic Appendix 2 Electronic Appendix 3 Electronic Appendix 4 Electronic Appendix 5 Electronic Appendix 6 References

 

	Electronic Appendix 2

Top Abstract Materials and Methods Results Discussion Discussion Electronic Appendix 1 Electronic Appendix 2 Electronic Appendix 3 Electronic Appendix 4 Electronic Appendix 5 Electronic Appendix 6 References

	Electronic Appendix 3

Top Abstract Materials and Methods Results Discussion Discussion Electronic Appendix 1 Electronic Appendix 2 Electronic Appendix 3 Electronic Appendix 4 Electronic Appendix 5 Electronic Appendix 6 References

	Electronic Appendix 4

Top Abstract Materials and Methods Results Discussion Discussion Electronic Appendix 1 Electronic Appendix 2 Electronic Appendix 3 Electronic Appendix 4 Electronic Appendix 5 Electronic Appendix 6 References

	Electronic Appendix 5

Top Abstract Materials and Methods Results Discussion Discussion Electronic Appendix 1 Electronic Appendix 2 Electronic Appendix 3 Electronic Appendix 4 Electronic Appendix 5 Electronic Appendix 6 References

	Electronic Appendix 6

Top Abstract Materials and Methods Results Discussion Discussion Electronic Appendix 1 Electronic Appendix 2 Electronic Appendix 3 Electronic Appendix 4 Electronic Appendix 5 Electronic Appendix 6 References

	Footnotes

 
Drs McCarthy and Gillinov are consultants to Atricure, Inc, West Chester, Ohio.

	References

Top Abstract Materials and Methods Results Discussion Discussion Electronic Appendix 1 Electronic Appendix 2 Electronic Appendix 3 Electronic Appendix 4 Electronic Appendix 5 Electronic Appendix 6 References

 

1. Bando K, Kasegawa H, Okada Y. The impact of pre- and postoperative atrial fibrillation on outcome after mitral valvuloplasty for nonischemic mitral regurgitation. J Thorac Cardiovasc Surg. 2004In press.
   
2. Bando K, Kobayashi J, Kosakai Y, Hirata M, Sasako Y, Nakatani S, et al. Impact of Cox maze procedure on outcome in patients with atrial fibrillation and mitral valve disease. J Thorac Cardiovasc Surg 2002;124:575-583.[Abstract/Free Full Text]
   
3. Cox JL. Intraoperative options for treating atrial fibrillation associated with mitral valve disease. J Thorac Cardiovasc Surg 2001;122:212-215.[Free Full Text]
   
4. McCarthy PM, Gillinov AM, Castle L, Chung M, Cosgrove 3rd D. The Cox-Maze procedure. the Cleveland Clinic experience. Semin Thorac Cardiovasc Surg 2000;12:25-29.[Medline]
   
5. Ad N, Cox JL. Combined mitral valve surgery and the maze III procedure. Semin Thorac Cardiovasc Surg 2002;14:206-209.[Medline]
   
6. Damiano Jr RJ. Alternative energy sources for atrial ablation. judging the new technology. Ann Thorac Surg 2003;75:329-330.[Free Full Text]
   
7. Gillinov AM, Blackstone EH, McCarthy PM. Atrial fibrillation. current surgical options and their assessment. Ann Thorac Surg 2002;74:2210-2217.[Abstract/Free Full Text]
   
8. Gillinov AM, McCarthy PM. AtriCure bipolar radiofrequency clamp for intraoperative ablation of atrial fibrillation. Ann Thorac Surg 2002;74:2165-2168.[Abstract/Free Full Text]
   
9. Prasad SM, Maniar HS, Schuessler RB, Damiano Jr RJ. Chronic transmural atrial ablation by using bipolar radiofrequency energy on the beating heart. J Thorac Cardiovasc Surg 2002;124:708-713.[Abstract/Free Full Text]
   
10. Sie HT, Beukema WP, Elvan A, Ramdat Misier AR. Long-term results of irrigated radiofrequency modified maze procedure in 200 patients with concomitant cardiac surgery. six years experience. Ann Thorac Surg 2004;77:512-517.[Abstract/Free Full Text]
    
11. Mohr FW, Fabricius AM, Falk V, Autschbach R, Doll N, Von Oppell U, et al. Curative treatment of atrial fibrillation with intraoperative radiofrequency ablation. short-term and midterm results. J Thorac Cardiovasc Surg 2002;123:919-927.[Abstract/Free Full Text]
    
12. Haissaguerre M, Jais P, Shah DC, Takahashi A, Hocini M, Quiniou G, et al. Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J Med 1998;339:659-666.[Abstract/Free Full Text]
    
13. Fuster V, Ryden LE, Asinger RW, Cannom DS, Crijns HJ, Frye RL, et al. ACC/AHA/ESC guidelines for the management of patients with atrial fibrillation. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines and Policy Conferences (Committee to develop guidelines for the management of patients with atrial fibrillation) developed in collaboration with the North American Society of Pacing and Electrophysiology. Eur Heart J 2001;22:1852-1923.[Free Full Text]
    
14. Blackstone EH, Naftel DC, Turner Jr ME. The decomposition of time-varying hazard into phases, each incorporating a separate stream of concomitant information. J Am Stat Assoc 1986;81:615-624.
    
15. Breiman L. Bagging predictors. Machine Learning 1996;24:123-140.
    
16. Blackstone EH. Breaking down barriers. helpful breakthrough statistical methods you need to understand better. J Thorac Cardiovasc Surg 2001;122:430-439.[Free Full Text]
    
17. Obadia JF, el Farra M, Bastien OH, Lievre M, Martelloni Y, Chassignolle JF. Outcome of atrial fibrillation after mitral valve repair. J Thorac Cardiovasc Surg 1997;114:179-185.[Abstract/Free Full Text]
    
18. Kobayashi J, Kosakai Y, Nakano K, Sasako Y, Eishi K, Yamamoto F. Improved success rate of the maze procedure in mitral valve disease by new criteria for patients’ selection. Eur J Cardiothorac Surg 1998;13:247-252.[Abstract/Free Full Text]
    
19. Scherer M, Dzemali O, Aybek T, Wimmer-Greinecker G, Moritz A. Impact of left atrial size reduction on chronic atrial fibrillation in mitral valve surgery. J Heart Valve Dis 2003;12:469-474.[Medline]
    
20. Zembala M, Lenarczyk R, Kalarus Z, Puszczewicz D, Przybylski R, Pacholewicz J. Early and late outcome after microwave ablation for chronic valvular atrial fibrillation. Heart Surg Forum 2003;6:403-408.[Medline]
    
21. Chen MC, Chang JP, Guo GB, Chang HW. Atrial size reduction as a predictor of the success of radiofrequency maze procedure for chronic atrial fibrillation in patients undergoing concomitant valvular surgery. J Cardiovasc Electrophysiol 2001;12:867-874.[Medline]
    
22. Kosakai Y, Kawaguchi AT, Isobe F, Sasako Y, Nakano K, Eishi K, et al. Cox maze procedure for chronic atrial fibrillation associated with mitral valve disease. J Thorac Cardiovasc Surg 1994;108:1049-1055.[Abstract/Free Full Text]
    
23. Haines DE. With pulmonary vein isolation for paroxysmal atrial fibrillation ablation, one size does not fit all. J Cardiovasc Electrophysiol 2002;13:962-963.[Medline]
    
24. Luria DM, Nemec J, Etheridge SP, Compton SJ, Klein RC, Chugh SS, et al. Intra-atrial conduction block along the mitral valve annulus during accessory pathway ablation. evidence for a left atrial "isthmus.". J Cardiovasc Electrophysiol 2001;12:744-749.[Medline]
    
25. Cox JL, Ad N. The importance of cryoablation of the coronary sinus during the maze procedure. Semin Thorac Cardiovasc Surg 2000;12:20-24.[Medline]
    
26. Deneke T, Khargi K, Grewe PH, von Dryander S, Kuschkowitz F, Lawo T, et al. Left atrial versus bi-atrial Maze operation using intraoperatively cooled-tip radiofrequency ablation in patients undergoing open-heart surgery. safety and efficacy. J Am Coll Cardiol 2002;39:1644-1650.[Abstract/Free Full Text]
    
27. Gaynor SL, Schuessler RB, Bailey MS, Ishii Y, Boineau JP, Gleva MJ, et al. Surgical treatment of atrial fibrillation. predictors of late recurrence. J Thorac Cardiovasc Surg 2005;129:104-111.[Abstract/Free Full Text]
    
28. Pappone C, Santinelli V, Manguso F, Vicedomini G, Gugliotta F, Augello G, et al. Pulmonary vein denervation enhances long-term benefit after circumferential ablation for paroxysmal atrial fibrillation. Circulation 2004;109:327-334.[Abstract/Free Full Text]
    
29. Oral H, Scharf C, Chugh A, Hall B, Cheung P, Good E, et al. Catheter ablation for paroxysmal atrial fibrillation. segmental pulmonary vein ostial ablation versus left atrial ablation. Circulation 2003;108:2355-2360.[Abstract/Free Full Text]
    
30. Marrouche NF, Dresing T, Cole C, Bash D, Saad E, Balaban K, et al. Circular mapping and ablation of the pulmonary vein for treatment of atrial fibrillation. impact of different catheter technologies. J Am Coll Cardiol 2002;40:464-474.[Abstract/Free Full Text]
    
31. Pacifico A, Henry PD. Ablation for atrial fibrillation. are cures really achieved?. J Am Coll Cardiol 2004;43:1940-1942.[Abstract/Free Full Text]
    

This article has been cited by other articles:

				G. S.C. Geuzebroek, P. K.E.W. Ballaux, N. M. van Hemel, J. C. Kelder, and J. J.A.M.T. Defauw Medium-term outcome of different surgical methods to cure atrial fibrillation: is less worse? Interactive CardioVascular and Thoracic Surgery, April 1, 2008; 7(2): 201 - 206. [Abstract] [Full Text] [PDF]

				F. Onorati, A. Esposito, G. Messina, A. di Virgilio, and A. Renzulli Right Isthmus Ablation Reduces Supraventricular Arrhythmias After Surgery for Chronic Atrial Fibrillation Ann. Thorac. Surg., January 1, 2008; 85(1): 39 - 48. [Abstract] [Full Text] [PDF]

				M. A. Groh, O. A. Binns, H. G. Burton III, S. W. Ely, and A. M. Johnson Ultrasonic Cardiac Ablation for Atrial Fibrillation During Concomitant Cardiac Surgery: Long-Term Clinical Outcomes Ann. Thorac. Surg., December 1, 2007; 84(6): 1978 - 1983. [Abstract] [Full Text] [PDF]

				A.M. Patwardhan Intraoperative ablation of atrial fibrillation using bipolar output of surgical radiofrequency generator (diathermy) and reusable bipolar forceps J. Thorac. Cardiovasc. Surg., June 1, 2007; 133(6): 1683 - 1683. [Full Text] [PDF]

				H. Calkins, J. Brugada, D. L. Packer, R. Cappato, S.-A. Chen, H. J.G. Crijns, R. J. Damiano Jr, D. W. Davies, D. E. Haines, M. Haissaguerre, et al. HRS/EHRA/ECAS Expert Consensus Statement on Catheter and Surgical Ablation of Atrial Fibrillation: Recommendations for Personnel, Policy, Procedures and Follow-Up: A report of the Heart Rhythm Society (HRS) Task Force on Catheter and Surgical Ablation of Atrial Fibrillation Developed in partnership with the European Heart Rhythm Association (EHRA) and the European Cardiac Arrhythmia Society (ECAS); in collaboration with the American College of Cardiology (ACC), American Heart Association (AHA), and the Society of Thoracic Surgeons (STS). Endorsed and Approved by the governing bodies of the American College of Cardiology, the American Heart Association, the European Cardiac Arrhythmia Society, the European Heart Rhythm Association, the Society of Thoracic Surgeons, and the Heart Rhythm Society. Europace, June 1, 2007; 9(6): 335 - 379. [Full Text] [PDF]