Share this article on:

Surgical Ablation for Atrial Fibrillation in Cardiac Surgery: A Meta-Analysis and Systematic Review

Cheng, Davy C. H. MD*; Ad, Niv MD†; Martin, Janet PharmD, MSc(HTA&M)*‡; Berglin, Eva E. MD, PhD§; Chang, Byung-Chul MD¶; Doukas, George MD∥; Gammie, James S. MD**; Nitta, Takashi MD, PhD††; Wolf, Randall K. MD‡‡; Puskas, John D. MD§§

Innovations: Technology & Techniques in Cardiothoracic & Vascular Surgery: March/April 2010 - Volume 5 - Issue 2 - pp 84-96
doi: 10.1097/IMI.0b013e3181d9199b
Original Article

Objectives: This meta-analysis sought to determine whether surgical ablation improves clinical outcomes and resource utilization compared with no ablation in adult patients with persistent and permanent atrial fibrillation (AF) undergoing cardiac surgery.

Methods: A comprehensive search was undertaken to identify all randomized (RCT) and nonrandomized (non-RCT) controlled trials of surgical ablation versus no ablation in patients with AF undergoing cardiac surgery up to April 2009. The primary outcome was sinus rhythm. Secondary outcomes included survival and any other reported clinically relevant outcome or indicator of resource utilization. Odds ratios (OR) and weighted mean differences (WMD) and their 95% confidence intervals (95% CI) were analyzed as appropriate using the random effects model. Heterogeneity was measured using the I2 statistic. Meta-regression was performed to explore the relationship between the benefit from surgical AF and duration of follow-up.

Results: Thirty-three studies met the inclusion criteria (10 RCTs and 23 non-RCTs) for a total of 4647 patients. The number of patients in sinus rhythm was significantly improved at discharge in the surgical AF ablation group versus (68.6%) the surgery alone group (23.0%) in RCTs (OR 10.1, 95% CI 4.5–22.5) and non-RCTs (OR 7.15, 95% CI 3.42–14.95). This effect on sinus rhythm (74.6% vs. 18.4%) remained at follow-up of 1 to 5 years (OR 6.7, 95% CI 2.8–15.7 for RCT, and OR 15.5, 95% CI 6.6–36.7 for non-RCT). The risk of all-cause mortality at 30 days was not different between the groups in RCT (OR 1.20, 95% CI 0.52–3.16) or non-RCT studies (OR 0.99, 95% CI 0.52–1.87). In studies reporting all-cause mortality at 1 year or more (up to 5 years), mortality did not differ in RCT studies (OR 1.21, 95% CI 0.59–2.51) but was significantly reduced in non-RCT studies (OR 0.54, 95% CI 0.31–0.96). Stroke incidence was not reduced significantly; however, in meta-regression, the risk of stroke decreased significantly with longer follow-up. Other clinical outcomes were similar between groups. Operation time was significantly increased with surgical AF ablation; however, overall impact on length of stay was variable.

Conclusions: In patients with persistent or permanent AF who present for cardiac surgery, the addition of surgical AF ablation led to a significantly higher rate of sinus rhythm in RCT and non-RCT studies compared with cardiac surgery alone, and this effect remains robust over the longer term (1–5 years). Although non-RCT studies suggest the possibility of reduced risk of stroke and death, this remains to be proven in prospective RCTs with adequate power and follow-up.

From the *Department of Anesthesia and Perioperative Medicine, Evidence-Based Perioperative Clinical Outcomes Research Group (EPiCOR), London Health Sciences Centre, The University of Western, Ontario, London, ON, Canada; †Department of Cardiac Surgery, Inova Heart and Vascular Institute, Falls Church, VA USA; ‡High Impact Technology Evaluation Centre, London Health Sciences Centre, London, ON, Canada; §Department of Cardiothoracic Surgery, Sahlgrenska University Hospital, Goteborg, Sweden; ¶Department of Cardiac Surgery, Yonsei University College of Medicine, Seoul, Korea; ∥Department of Cardiothoracic Surgery, Glenfield Hospital, Leicester, UK; **Division of Cardiac Surgery, University of Maryland School of Medicine, Baltimore, MD USA; ††Department of Cardiac Surgery, Nippon Medical School Main Hospital, Tokyo, Japan; ‡‡Deaconess Hospital, Cincinnati, OH USA; and §§Division of Cardiothoracic Surgery, Emory University, Atlanta, GA USA.

Accepted for publication February 11, 2010.

Support for the meta-analysis and a consensus conference was provided by the International Society for Minimally Invasive Cardiothoracic Surgery (ISMICS), which has received unrestricted educational grants from industries that produce surgical technologies.

Disclosures: Eva E. Berglin, MD, PhD is a consultant for ATS Medical, Inc., Minneapolis, MN USA. James S. Gammie, MD is a consultant for ATS Medical, Inc., Minneapolis, MN USA. Randall K. Wolf, MD is a consultant and receives royalties from AtriCure, West Chester, OH USA. John D. Puskas, MD is a consultant for MAQUET, Wayne, NJ USA, and Medtronic, Inc., Minneapolis, MN USA.

Address correspondence and reprint requests to Davy C. H. Cheng, MD, LHSC-University Hospital, 339 Windermere Road, C3-172, London, ON, Canada N6A 5A5. E-mail:

Annual stroke risk in patients with nonrheumatic atrial fibrillation (AF) not taking anticoagulation therapy is 4.5%.1 Although antiarrhythmic therapies may be used for rhythm or rate control,2 these medications are not universally effective, and chronic requirement for medication and anticoagulation may also adversely affect quality of life (QOL).

Patients with mitral valve and/or coronary artery disease who are having AF and are candidates for cardiac surgery for valve repair/replacement or for coronary artery bypass may be particularly suited to surgical AF ablation. The Cox maze procedure was first performed in St. Louis in 1987 with the intention of eliminating AF by using incision scars to block the abnormal electrical circuits (atrial macroreentry) that are characteristic of AF. This procedure requires an extensive series of endocardial incisions through both atria while on cardiopulmonary bypass, with access to the heart via median sternotomy.3 A series of systematic improvements have subsequently been made, culminating in the 1992 Cox maze III procedure, which is now considered to be the “gold standard” for effective surgical treatment of AF. The Cox maze III is sometimes referred to as the traditional maze, the cut-and-sew maze, or simply the maze.4 Further modifications of surgical AF ablation in subsequent years have focused on using various energy modalities to create lesion sets with fewer cardiac incisions. These have included radiofrequency, ultrasound, microwave, laser, and cryotherapy.

A comprehensive and methodologically rigorous meta-analysis of all comparative trials (randomized [RCT] and nonrandomized trials [non-RCT]) is needed to better weigh the risks and benefits of surgical ablation when added to other planned cardiac surgery. Previous meta-analyses and systematic reviews have evaluated only RCTs, which may underestimate the risks if underpowered or of short duration.5–7 This systematic review and meta-analysis will attempt to address the totality of evidence available to date.

Back to Top | Article Outline


This systematic review with meta-analysis sought to determine whether, in patients with AF undergoing cardiac surgery, AF ablation improves clinical outcomes, QOL, and resource-related outcomes when compared with cardiac surgery without ablation. To answer this primary objective, the following subquestions were prespecified to guide the systematic review:

1. Does ablation improve achievement of normal sinus rhythm (NSR) before discharge, and is the effect sustained long-term (6 months, 1 year, 3 years, 5 years, and 10 years)?

2. Does ablation reduce the need for pharmacologic treatment of AF, direct current cardioversion, or pacemaker insertion?

3. Does ablation reduce the risk of stroke, myocardial infarction, heart failure, and other complications?

4. Does ablation improve QOL, functionality, or other patient-reported outcomes?

5. Does ablation reduce total costs, intensive care unit (ICU) and hospital length of stay, need for repeat cardiac surgery, and readmissions, and is it cost-effective?

The secondary objective of this systematic review was also defined as follows: in patients with AF undergoing cardiac surgery plus ablation, how do different ablative techniques compare with each other for conversion to NSR (cut-and-sew, Cox mazes I, II, and III, pulmonary vein (PV) isolation, radiofrequency ablation, ultrasound ablation, laser ablation, and cryoablation)?

Back to Top | Article Outline


This meta-analysis of comparative studies was performed in accordance with state of the art methodological recommendations (ie, as per QUOROM and MOOSE guidelines)8,9 and according to a protocol that prespecified outcomes, search strategies, inclusion criteria, and statistical analyses.

Back to Top | Article Outline

Definition of Endpoints

Clinical endpoints of interest included achievement of sinus rhythm, need for antiarrhythmic drugs, direct current cardioversion, need for pacemaker, reexploration for bleeding, major bleeding, transfusions, all-cause mortality, stroke or transient ischemic attack (TIA), thromboembolic events, myocardial infarction, low cardiac outcome, renal failure, heart failure, New York Heart Association (NYHA) class, reintervention for bleeding, reintervention for paravalvular leak, patient satisfaction, functionality, QOL, cross-clamp time, duration of surgery, ICU length of stay, total hospital length of stay, costs, and cost-effectiveness. For this analysis, AF was defined as recommended by the American College of Cardiology, American Heart Association, and the European Society of Cardiology guidelines, given in Table 1.10

Back to Top | Article Outline

Literature Search

A comprehensive literature search of MEDLINE, EMBASE, Cochrane CENTRAL, Current Contents, and Science Citation Index, using keywords and variants of, was performed from the earliest available date to April 2009. The most recent 6 months of relevant surgical and anesthesia journals were searched by hand, and databases of conference abstracts were reviewed electronically. Experts were contacted to solicit additional reports of published or unpublished clinical trials of surgical AF ablation, as negative studies are more likely to be unpublished.

Back to Top | Article Outline

Inclusion Criteria

To be eligible for inclusion in the systematic review and meta-analysis, trials had to be RCTs or non-RCTs comparing surgical AF ablation performed during cardiac surgery versus cardiac surgery alone in patients undergoing valvular or bypass surgery. All surgical AF ablation techniques were eligible for inclusion, including the Cox maze procedure and its variants (cut and sew), or ablation using other modalities such as radiofrequency ablation, cryoablation, lasers, microwaves, whether applied biatrially or on a single atrium, with or without pulmonary vein isolation. Subanalyses by type of procedure and technology was planned to evaluate relative efficacy. Studies of patients with lone AF and those undergoing catheter-based ablative procedures were excluded.

Back to Top | Article Outline

Data Extraction

Two authors (D.C. and J.M.) independently extracted the following data points: baseline demographics including number of patients, inclusion/exclusion criteria for patient, and type of AF. Details of the ablative procedure and the comparative conventional treatment provided were recorded. Two authors extracted outcomes data and verified the extraction with each other. Discrepancies were resolved by consensus. No unpublished study was found.

Back to Top | Article Outline

Data Analysis

Odds ratios (OR) and their 95% confidence intervals (95% CI) were calculated for discrete data. Weighted mean differences (WMD, 95% CI) were calculated for continuous data when similar metrics were used across the reported outcomes. Heterogeneity was explored through the Q-statistic and by calculating the I2. Summary ORs and WMDs were calculated using the random effects model to provide a conservative analysis in the case of heterogeneity. When statistical heterogeneity was high (I2 >50%), reasons for heterogeneity were explored. When possible, data were analyzed by intention-to-treat approach. When dropouts from the clinical trials were unaccounted for, we planned to do worst-case scenario sensitivity analyses to determine whether the conclusions were robust. Meta-regression was performed to assess the relationship between effect size and length of follow-up. Statistical significance for overall effect was defined as P < 0.05 or a confidence that excluded the value 1.00 for ORs and 0 for WMDs.

Back to Top | Article Outline


Study Identification

Of >4303 citations screened, 475 were identified as potentially relevant and were retrieved for review. Of these, 33 met the inclusion criteria for the primary analysis of surgical AF ablation (10 RCTs including a total 650 patients11–21 and 23 non-RCTs of 2997 patients,22–46 for an overall total of 4647 patients; Fig. 1, QUOROM Flowchart). For trials that reported more than once on an identical or overlapping population, only the most recent or most complete results were included. Some non-RCTs may have excluded patients with early complications or if ablative surgery was aborted, and in many non-RCTs, patient attrition was inadequately described. Publication bias was not evident after visual inspection of funnel plots; however, this test was underpowered for most clinical outcomes.

Back to Top | Article Outline

Patient Characteristics

Baseline characteristics are listed in Table 2. At baseline, the mean age was 65 years, and ∼50% were male. Although baseline patient characteristics were similar for RCT and non-RCT, the small size of the RCT and the inconsistent reporting of baseline characteristics in non-RCT precluded adequate power to rule out potential differences. More than 85% of patients had permanent or persistent AF, and ∼15% had paroxysmal AF.

Back to Top | Article Outline


Significant heterogeneity across studies was observed for achievement of sinus rhythm, cardioversion, cardiac death, continuation of antiarrhythmics, duration of surgery, and length of stay. Nevertheless, for most of these outcomes, the heterogeneity was less concerning because it was driven by uncertainty about the size of effect and not the direction of effect and, therefore, did not generally put into question the presence of significant benefit (when detected), with the exception of antiarrhythmics, cardiac death, and cardioversion.

Back to Top | Article Outline

Sinus Rhythm

The number of patients discharged from hospital in sinus rhythm was significantly higher for the surgical AF ablation group (68.6%) versus the non-AF ablation group (23.0%) in RCTs (OR 010.1, 95% CI 4.5–22.5, 8 RCTs) and non-RCTs (OR 7.15, 95% CI 3.42–14.95, 12 non-RCTs; Fig. 2A). This significant effect on sinus rhythm was maintained at 1 year in RCTs (OR 10.9, 95% CI 5.4–22.1, 8 RCTs) and in non-RCTs (OR 9.53, 95% CI 5.5–16.6, 10 non-RCTs). In trials that reported longer term results (1–5 years), there were significantly more patients in sinus rhythm in the ablation group (74.6% vs. 18.4%; OR 6.7, 95% CI 2.8–15.7; 1 RCT and OR 15.5, 95% CI 6.6–36.7, 10 non-RCTs; Fig. 2B). The number of patients undergoing direct current cardioversion was not different between the groups (Table 3).

Back to Top | Article Outline

Antiarrhythmics and Pacemakers

There was significant heterogeneity in the trials regarding the continuation of antiarrhythmics post-AF ablation. Some trials reported routine long-term continuation regardless of achievement of sinus rhythm, whereas other studies did not report information on local policies regarding antiarrhythmic continuation. Although antiarrhythmic drug use was not significantly reduced in the surgical AF ablation group versus nonablation group in RCTs (OR 0.96, 95% CI 0.33–2.82; 5 RCTs), there was significant reduction in antiarrhythmic drug use in non-RCTs (OR 0.17, 0.07–0.37). However, the majority of the heterogeneity was driven by two RCTs in which policies specified that antiarrhythmics should be routinely continued.

The number of patients with AV block was not significantly different between groups in RCT (OR 0.98, 95% CI 0.20–4.91; 2 RCTs) and non-RCT studies (OR 2.39, 95% CI 0.67–8.52; 3 non-RCTs). The number of patients requiring a temporary pacemaker did not differ between the groups. Similarly, the number of patients requiring permanent pacemaker insertion was not different in RCT (OR 1.2, 95% CI 0.5–2.6; 7 RCTs) or in non-RCT studies (OR 1.9, 95% CI 0.9–3.7; 11 non-RCTs; Fig. 3).

Back to Top | Article Outline


The risk of 30-day all-cause mortality was not different between the groups in RCT (OR 1.20, 95% CI 0.52–3.16; 7 RCTs) or non-RCT studies (OR 0.99, 95% CI 0.52–1.87; 12 non-RCTs). All-cause mortality was not significantly different at 1 year in RCT (OR 1.07, 95% CI 0.47–2.43; 7 RCTs) or non-RCT studies (OR 1.04, 95% CI 0.45–2.43; 5 non-RCTs; Fig. 4A). In studies reporting all-cause mortality at 2 years or more (up to 5 years), all-cause mortality did not differ in RCT studies (OR 1.92, 95% CI 0.41–9.04; 8 RCTs) but was significantly reduced after surgical ablation of AF in non-RCT studies (OR 0.41, 95% CI 0.21–0.79; 10 non-RCTs; Fig. 4B). The risk of cardiac death was similar between the groups in RCT and non-RCT studies.

Back to Top | Article Outline

Stroke and Thromboembolic Events

The risk of stroke or TIA in hospital did not differ between the groups for RCT (OR 0.71, 95% CI 0.06–8.36; 2 RCTs) or non-RCT studies (OR 0.68, 95% CI 0.15–2.98; 2 non-RCTs). When all trials reporting stroke at follow-up were considered, there was no significant difference between the groups for RCT studies (OR 0.92, 95% CI 0.20–4.32; 4 RCTs); however, there was a significant reduction in the risk of stroke or TIA in favor of ablation in non-RCT studies (OR 0.25, 95% CI 0.14–0.46; 10 RCTs). When stroke or thromboembolic events were considered in aggregate, the difference between groups did not reach significance for RCT studies (OR 0.35, 95% CI 0.09–1.32, 5 RCTs), but it reached significance for non-RCT studies (OR 0.28, 95% CI 0.15–0.52; 10 non-RCTs) in favor of ablation (Fig. 5A).

Back to Top | Article Outline

Length of Follow-Up and Effect on Stroke

When meta-regression revealed a statistically significant relationship between stroke reduction and time since surgery, it suggests that surgical AF ablation provided increasing protection from stroke with increased follow-up (P = 0.03; Fig. 5B). Similarly, for the aggregate of stroke or thromboembolic events, surgical AF ablation provided significantly greater protection as the length of follow-up increased (P = 0.007).

Back to Top | Article Outline

Other Clinical Outcomes

The risk of myocardial infarction, low cardiac output syndrome, intraaortic balloon pump, bleeding events, cardiac tamponade, reintervention for valvular leak or regurgitation, renal dysfunction, and congestive heart failure did not differ between groups for RCT or non-RCT studies. Although reoperation for bleeding did not differ in RCTs (OR 0.61, 95% CI 0.19–1.92; 4 RCTs), there was a significant increase in risk of reoperation for bleeding after ablation of AF in non-RCT studies (OR 2.63, 95% CI 1.11–6.22; 7 non-RCTs).

Back to Top | Article Outline

Patient Functionality Outcomes and QOL

Left ventricular ejection fraction was significantly improved after ablation of AF in RCT studies (WMD 4%, 95% CI 1%–2%, 3 RCTs) but not in non-RCT studies (WMD −2, 95% CI +1% to −5%; 2 non-RCTs). The number of patients remaining in NYHA classes III to IV was not different between the groups in RCT or non-RCT studies.

Exercise testing was performed in two RCTs. De Lima et al14 reported that the chronotropic response was significantly better in the surgical ablation group (pulmonary vein isolation [PVI] or maze) compared with control (86.2% or 80.5% for ablation vs. 64.1% for control, P < 0.05); however, in this study, only NSR patients underwent testing. Deneke et al15 reported that although VO2 max was not different between groups, the maximum workload was increased for maze versus control in 22 patients (P = 0.008). Doukas et al16 reported that patients undergoing surgical ablation recorded longer distances than control patients on the shuttle walk test (P = 0.02); although, when the increase in distance over baseline was compared, the significance did not remain (P = 0.13).

QOL was reported in two RCTs. Jessurun et al18 reported no significant difference between surgical AF ablation and control patients on Short Form-36 domains at 3 months and 12 months. Similarly, von Oppell21 found that Short Form-36 scores generally did not differ between the groups.

Back to Top | Article Outline

Resource-Related Outcomes

Cardiopulmonary bypass time was significantly greater by a mean of 27 to 55 minutes in the surgical AF ablation group compared with surgery alone in the RCT (WMD 27 minutes, 95% CI 5–16 minutes; 8 RCTs) and in the non-RCT studies (WMD 55 minutes, 95% CI 6–43 minutes; 13 non-RCTs). Cross-clamp time was also significantly greater by a mean of 15 to 41 minutes in RCT (WMD 15 minutes, 95% CI 4–7 minutes; 6 RCTs) and in non-RCT studies (WMD 41 minutes, 95% CI 5–30 minutes; 11 non-RCTs). Operation time was significantly greater for surgical AF ablation in non-RCT (WMD 166 minutes, 95% CI 14–139 minutes; 2 non-RCTs) but not in the sole RCT reporting this outcome (WMD 7 minutes, 95% CI −4 to 18 minutes; 1 RCT). Length of stay in ICU was not reduced significantly in the sole RCT reporting this outcome (WMD −0.5, 95% CI −1.5 to 0.5; 1 RCT) and was significantly increased in non-RCT studies (WMD 0.9 days, 95% CI 0.3–1.6 days; 4 non-RCTs). Total hospital length of stay was not significantly different between groups in RCT studies (WMD 0.7 days, 95% CI −0.8 to 2.2 days, 5 RCTs) but was significantly increased after ablation of AF in non-RCT studies (WMD 1.5 days, 95% CI 0.6–2.3, 5 non-RCT studies; Table 4).

Back to Top | Article Outline

Subanalyses by Type of Procedure or Modality

Subanalyses by type of procedure (cut-and-sew vs. energy ablation) did not show significant differences for success with different procedures (ie, the relative odds for achievement of sinus rhythm for surgical ablation over no ablation was similar across different modalities of ablation). There was insufficient information to evaluate differences by pulmonary vein isolation versus biatrial versus single atrial procedures.

Back to Top | Article Outline


Overall, this meta-analysis suggests that surgical AF ablation results in better rates of sinus rhythm both early and late (≥1 year) after surgery compared with patients not having surgical ablation. The results obtained from the RCT and non-RCT studies are largely congruent with each other regarding the increase in number of patients achieving sinus rhythm with surgical ablation. More than 70% of patients in the surgical ablation group were in sinus rhythm at 1 year, compared with ∼20% in the surgery only group. Although statistically significant heterogeneity was found, this heterogeneity was because of the magnitude of reduction varying across trials and was not because of differences in direction of effect across studies (all studies showed improved likelihood of achieving sinus rhythm after surgical ablation of AF, with ORs ranging from 3 to 77).

Whether surgical AF ablation allows more patients to discontinue antiarrhythmics remains uncertain because of the relatively few trials that reported this outcome and because of the heterogeneity across trials in reporting us of antiarrhythmic drugs. Some studies reported antiarrhythmics by class, whereas others reported antiarrhythmics and rate control drugs only in aggregate. Also, some studies routinely continued rhythm and rate control drugs in patients regardless of achievement of sinus rhythm, whereas other studies left the decisions to the discretion of prescribers. It is important to note that none of the trials included in this meta-analysis were prospectively powered to measure the ability to discontinue antiarrhythmics and rate control drugs. Although there was significant reduction in the use of antiarrhythmics in non-RCTs, this was not true for RCTs. However, the latter included two RCTs, which specified that antiarrhythmics should be continued postprocedure regardless of rhythm. Similarly, for oral anticoagulants (not assessed in this meta-analysis), there was inconsistent reporting of policies and prescribing patterns for patients who achieved sinus rhythm postprocedure.

Permanent pacemaker insertion was also heterogeneous across trials, and although there was no difference in RCTs, there was a trend toward increased incidence of pacemaker insertion in the surgical AF ablation group in non-RCT studies. This raises concern about the risk of iatrogenically induced bradycardia or other arrhythmias during or after AF ablation. Because of the long-term implications, heterogeneous data, and relatively few RCTs reporting this outcome, further study will be necessary to better characterize whether pacemaker insertion is truly an increased risk with surgical ablation (and, if so, whether it is associated with different techniques used to surgically ablate AF such as with pulmonary vein isolation, with cut-and-sew, or with certain mapping techniques). Retrospective evaluation of multicenter experience in the United States from the Society of Thoracic Surgeons database suggests that the risk of pacemaker insertion is significantly increased with surgical AF ablation compared with surgery alone.47

Overall, mortality was not affected by applying surgical ablation when compared with the nonablation group both for the RCT and non-RCTs up to 1 year. However, it is notable that all-cause mortality was significantly lower in the surgical ablation groups in the non-RCTs with longer term follow-up (mean all-cause mortality was 5.4% vs. 13.4% in the ablation vs. control groups at midterm follow-up of 1–5 years).

Stroke and TIAs were also reduced after surgical ablation of AF in non-RCT studies but not in RCTs. However, only two RCTs reported strokes. When all studies were analyzed by length of follow-up, the strength of association magnitude of stroke reduction in the surgical AF ablation group improved over time compared with the control group, suggesting that there is greater reduction in stroke with longer term follow-up. Similarly, for stroke and thromboembolic events considered in aggregate, there was a significant reduction in the surgical AF ablation group, and this effect was strengthened with longer term follow-up.

Other outcomes, including NYHA class, ejection fraction, exercise testing, and QOL scores were not significantly impacted by surgical AF ablation; however, these outcomes were poorly reported, and in general, the direction of effect showed a trend toward a benefit in favor of ablation.

As expected, cross-clamp times and operation time were significantly increased with surgical AF ablation. However, there was significant heterogeneity across trials as varied levels of experience and techniques were used. The increase in cross-clamp and procedure times is difficult to generalize to other settings but should be considered before adopting the procedure for routine practice. Overall impact on costs and length of stay remains largely unstudied. Non-RCT studies suggested that hospital length of stay may be increased with surgical AF ablation; however, RCTs did not show a significant difference. It is important to note that some of the non-RCTs included early operator experience and some included more complex cut-and-sew patterns.

Back to Top | Article Outline

Strengths and Limitations

This meta-analysis provides a comprehensive systematic review of all available comparative studies of surgical AF ablation for patients undergoing cardiac surgery. To date, few meta-analyses have been done to address this question, and the surgeons and their patients have had to make decisions without a clear understanding of the evidence base. This evidence base should be interpreted in light of the remaining uncertainties.

A number of limitations should be highlighted to facilitate appropriate interpretation of this meta-analysis. First and foremost, some of the data come from non-RCT studies, which may be susceptible to bias in patient selection and incomplete ascertainment of outcomes. As a result, RCT data should be preferred to the non-RCT data whenever possible in this analysis. However, the RCTs reported a narrower range of outcomes, on a smaller number of patients, and were generally of shorter follow-up than the collective evidence available in the non-RCT studies. As a result, the non-RCT data adds information regarding a broader range of outcomes, including adverse effects, and over a longer follow-up.

Additional limitations include the wide range of dates of the included studies and the presumed range of surgeon expertise represented. The included trials were published in the years ranging from 1996 to 2009. The surgeons' level of expertise was rarely specified in the included trials and leaves the reader with uncertainty whether the results will apply equally to contemporary situations, with newer technologies, and an overall increasing expertise in the techniques of surgical ablation. Furthermore, the role of appendage exclusion, per se, is often incompletely reported.

Back to Top | Article Outline

Clinical Implications

This analysis confirms the potential role of surgical AF ablation for achievement of sinus rhythm. It suggests that surgical ablation should be discussed as an option for patients with AF who present for cardiac surgery. Whether the procedure will allow them to discontinue antiarrhythmics, rate control drugs, or anticoagulation over the longer term remains unproven, although the potential exists. Discussing the possibility of reduced stroke and thromboembolic events over the longer term, but potential increased need for pacemaker insertion, will also be key in the discussions with patients to allow for informed decision making.

Back to Top | Article Outline

Which Ablative Technique Is Best?

Unfortunately, this analysis did not allow for statistical differentiation in outcomes among the various ablative techniques used. Through subgroup analysis, we tested indirectly whether one technique is better than another for achieving sinus rhythm or for survival; however, the subgroups were underpowered to adequately test the various techniques used because there were few techniques that were similar enough to be combined into subgroups. Future systematic reviews are encouraged to compare head-to-head trials of ablative techniques and technologies. To date, no single technique or ablative technology can be definitively recommended over another on the basis of comparative trials.

Back to Top | Article Outline

What This Study Adds to Current Literature

This review provides important information regarding the potential for protection from stroke and thromboembolic events over the longer term and, also, raises the issue of potential for increased need for pacemaker insertion with surgical AF ablation. This current systematic review represents an update of the earlier analysis also allows for comparison of RCT and non-RCT data.

Back to Top | Article Outline

Further Research Required

Examination of the limitations of this meta-analysis highlights the need to address the deficiencies inherent in the evidence base for surgical AF ablation. This meta-analysis aggregates the best available evidence, but it does not obviate the need for further RCTs; rather, this meta-analysis highlights the need for future RCTs of adequate power to measure important differences in survival, stroke, thromboembolic events, and patient symptoms and QOL. Cost-effectiveness analyses alongside long-term RCTs should be performed to inform future policies in this area as AF ablation increases operating time but may prevent future adverse events related to stroke and thromboembolism.

Back to Top | Article Outline


Evidence to date suggests that surgical AF ablation significantly improves the odds of achieving sinus rhythm by more than 10-fold and this improvement is retained beyond 1 year when compared with cardiac surgery without ablation in patients with persistent or permanent AF who present for valvular and/or coronary artery bypass surgery. Non-RCT evidence suggests that surgical AF ablation may also reduce the risk of stroke and thromboembolic events but may increase need for pacemaker insertion. Longer term follow-up from existing and future RCTs should be encouraged to ascertain whether this reduction in stroke is a robust finding and over what time course antiarrhythmics can be safely discontinued. Finally, prospective, RCT studies are urgently needed to better define the best lesion sets to be used in each particular patient population and to help determine the superior ablation technologies.

Back to Top | Article Outline


The authors thank Ms. Aurelie Alger and Ms. Elizabeth Chouinard for their professional assistance in organizing the consensus conference; Dr. Kathy Fang and Dr. Myoung Kim for data extraction and verification; and Jennifer Podeszwa DeOliveira and Erin Boyce for facilitating literature searches and retrieval.

Back to Top | Article Outline


1.Ruel M, Masters RG, Rubens FD, et al. Late incidence and determinants of stroke after aortic and mitral valve replacement. Ann Thorac Surg. 2004;78:77–84.
2.Wyse DG, Waldo AL, DiMarco JP, et al. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med. 2002;347:1825–1833.
3.Valencia MJ, Climent PVE, Marin OF, et al. The efficacy of scheduled cardioversion in atrial fibrillation. Comparison of two schemes of treatment: electrical versus pharmacological cardioversion. J Rev Esp Cardiol. 2002;55:113–120.
4.Cox JL, Schuessler RB, Lappas DG, Boineau JP. An 8 ½-year clinical experience with surgery for atrial fibrillation. Ann Surg. 1996;224:267–273; discussion 273–275.
5.Khargi K, Keyhan-Falsafi A, Hutten BA, et al. Surgical treatment of atrial fibrillation: a systematic review. Herzschr Electrophys. 2007;18:68–76.
6.Reston JT, Shuhaiber JH. Meta-analysis of clinical outcomes of maze-related surgical procedures for medically refractory atrial fibrillation. Eur J Cardiothorac Surg. 2005;28:724–730.
7.Barnett S, Ad N. Surgical ablation as treatment for the elimination of atrial fibrillation: a meta-analysis. J Thorac Cardiovasc Surg. 2006;131:102935.
8.Moher D, Cook DJ, Eastwood S, et al; for the QUOROM Group. Improving the quality of reports of meta-analyses of randomized controlled trials: the QUOROM statements. Lancet. 1999;354:1896–1900.
9.Stroup DR, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis of Observational Studies in Epidemiology (MOOSE) group. JAMA. 2000;283:2008–2012.
10.Fuster V, Rydén LE, Cannom DS, et al. ACC/AHA/ESC 2006 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 (Writing Committee to Revise the 2001 Guidelines for the Management of Patients With Atrial Fibrillation): developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. Circulation. 2006;114:e257–e354.
11.Abreu Filho CA, Lisboa LA, Dallan LA, et al. Effectiveness of the maze procedure using cooled-tip radiofrequency ablation in patients with permanent atrial fibrillation and rheumatic mitral valve disease. Circulation. 2005;112:I20–I25.
12.Akpinar B, Guden M, Sagbas E, et al. Combined radiofrequency modified maze and mitral valve procedure through a port access approach: early and mid-term results. Eur J Cardiothorac Surg. 2003;24:223–230.
13.Blomstrom-Lundqvist C, Johansson B, Berglin E, et al. A randomized double-blind study of epicardial left atrial cryoablation for permanent atrial fibrillation in patients undergoing mitral valve surgery: The SWEDish multicentre atrial fibrillation study (SWEDMAF). Eur Heart J. 2007;28:2902–2908. Lima GG, Kalil RA, Leiria TL, et al. Randomized study of surgery for patients with permanent atrial fibrillation as a result of mitral valve disease. Ann Thorac Surg. 2004;77:2089–2094; discussion 2094–2095.
15.Deneke T, Khargi K, Grewe PH, et al. Efficacy of an additional MAZE procedure using cooled-tip radiofrequency ablation in patients with chronic atrial fibrillation and mitral valve disease. A randomized, prospective trial. Eur Heart J. 2002;23:558–566.
16.Doukas G, Samani NJ, Alexiou C, et al. Left atrial radiofrequency ablation during mitral valve surgery for continuous atrial fibrillation: a randomized controlled trial. JAMA. 2005;294:2323–2329.
17.Khargi K, Deneke T, Haardt H, et al. Saline-irrigated, cooled-tip radiofrequency ablation is an effective technique to perform the maze procedure. Ann Thorac Surg. 2001;72:S1090–S1095.
18.Jessurun ER, van Hemel NM, Defauw JJ, et al. A randomized study of combining maze surgery for atrial fibrillation with mitral valve surgery. J Cardiovasc Surg (Torino). 2003;44:9–18.
19.Schuetz A, Schulze CJ, Sarvanakis KK, et al. Surgical treatment of permanent atrial fibrillation using microwave energy ablation: a prospective randomized clinical trial. Eur J Cardiothorac Surg. 2003;24:475–480; discussion 480.
20.Srivastava V, Kumar S, Javali S, et al. Efficacy of three different ablative procedures to treat atrial fibrillation in patients with valvular heart disease: a randomised trial. Heart Lung Circ. 2008;17:232–240.
21.von Oppell UO, Masani N, O'Callaghan P, et al. Mitral valve surgery plus concomitant atrial fibrillation ablation is superior to mitral valve surgery alone with an intensive rhythm control strategy. Eur J Cardiothorac Surg. 2009;35:641–650.
22.Bai BJ, Zhong ZH, Xin LP, et al. The clinical study of circumferential ablation around orifice of pulmonary vein by radiofrequency energy with chronic atrial fibrillation undergoing rheumatic valvular heart surgery. Zhonghua Wai Ke Za Zhi. 2006;44:946–948.
23.Bando K, Kasegawa H, Okada Y, et al. Impact of preoperative and postoperative atrial fibrillation on outcome after mitral valvuloplasty for nonischemic mitral regurgitation. J Thorac Cardiovasc Surg. 2005;129:1032–1040.
24.Bando K, Kobayashi J, Hirata M, et al. Early and late stroke after mitral valve replacement with a mechanical prosthesis: risk factor analysis of a 24-year experience. J Thorac Cardiovasc Surg. 2003;126:358–364.
25.Bando K, Kobayashi J, Kosakai Y, 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.
26.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.
27.Damiano RJ Jr, Gaynor SL, Bailey M, et al. The long-term outcome of patients with coronary disease and atrial fibrillation undergoing the cox maze procedure. J Thorac Cardiovasc Surg. 2003;126:2016–2021.
28.Forlani S, De Paulis R, Guerrieri Wolf L, et al. Conversion to sinus rhythm by ablation improves quality of life in patients submitted to mitral valve surgery. Ann Thorac Surg. 2006;81:863–867.
29.Gaita F, Gallotti R, Calo L, et al. Limited posterior left atrial cryoablation in patients with chronic atrial fibrillation undergoing valvular heart sugery. J Am Coll Cardiol. 2000;36:159–166.
30.Guang Y, Zhen-jie C, Yong LW, et al. Evaluation of clinical treatment of atrial fibrillation associated with rheumatic mitral valve disease by radiofrequency ablation. Eur J Cardiothorac Surg. 2002;21:249–254.
31.Handa N, Schaff HV, Morris JJ, et al. Outcome of valve repair and the cox maze procedure for mitral regurgitation and associated atrial fibrillation. J Thorac Cardiovasc Surg. 1999;118:628–635.
32.Itoh T, Okamoto H, Nimi T, et al. Left atrial function after Cox's maze operation concomitant with mitral valve operation. Ann Thorac Surg. 1995;60:354–359; discussion 359–360.
33.Jatene MB, Marcial MB, Tarasoutchi F, et al. Influence of the maze procedure on the treatment of rheumatic atrial fibrillation—evaluation of rhythm control and clinical outcome in a comparative study. Eur J Cardiothorac Surg. 2000;17:117–124.
34.Johansson B, Houltz B, Berglin E, et al. Short-term sinus rhythm predicts long-term sinus rhythm and clinical improvement after intraoperative ablation of atrial fibrillation. Europace. 2008;10:610–617.
35.Kawaguchi AT, Kosakai Y, Sasako Y, et al. Risks and benefits of combined maze procedure for atrial fibrillation associated with organic heart disease. J Am Coll Cardiol. 1996;28:985–990.
36.Kim HK, Kim YJ, Kim KI, et al. Impact of the maze operation combined with left-sided valve surgery on the change in tricuspid regurgitation over time. Circulation. 2005;112:I14–I19.
37.Knaut M, Tugtekin SM, Matschke K. Pulmonary vein isolation by microwave energy ablation in patients with permanent atrial fibrillation. J Card Surg. 2004;19:211–215.
38.Kobayashi J, Kosakai Y, Isobe F, et al. Rationale of the cox maze procedure for atrial fibrillation during redo mitral valve operations. J Thorac Cardiovasc Surg. 1996;112:1216–1221; discussion 1222.
39.Kobayashi J, Sasako Y, Bando K, et al. Eight-year experience of combined valve repair for mitral regurgitation and maze procedure. J Heart Valve Dis. 2002;11:165–171; discussion 171–172.
40.Louagie Y, Buche M, Eucher P, et al. Improved patient survival with concomitant cox maze III procedure compared with heart surgery alone. Ann Thorac Surg. 2009;87:440–446.
41.Mantovan R, Raviele A, Buja G, et al. Left atrial radiofrequency ablation during cardiac surgery in patients with atrial fibrillation. J Cardiovasc Electrophysiol. 2003;14:1289–1295.
42.Myrdko T, Sniezek-Maciejewska M, Rudzinski P, et al. Efficacy of intra-operative radiofrequency ablation in patients with permanent atrial fibrillation undergoing concomitant mitral valve replacement. Kardiol Pol. 2008;66:932–938; discussion 939–940.
43.Patwardhan AM, Dave HH, Tamhane AA, et al. Intraoperative radiofrequency microbipolar coagulation to replace incisions of maze III procedure for correcting atrial fibrillation in patients with rheumatic valvular disease. Eur J Cardiothorac Surg. 1997;12:627–633.
44.Raanani E, Albage A, David TE, et al. The efficacy of the Cox/maze procedure combined with mitral valve surgery: a matched control study. Eur J Cardiothorac Surg. 2001;19:438–442.
45.Tuinenburg AE, Van Gelder IC, Tieleman RG, et al. Mini-maze suffices as adjunct to mitral valve surgery in patients with preoperative atrial fibrillation. J Cardiovasc Electrophysiol. 2000;11:960–967.
46.Yuda S, Nakatani S, Kosakai Y, et al. Mechanism of improvement in exercise capacity after the maze procedure combined with mitral valve surgery. Heart. 2004;90:64–69.
47.Gammie JS, Haddad M, Milford-Feland S, et al. Atrial fibrillation correction surgery: lessons from the Society of Thoracic Surgeons National Cardiac Database. Ann Thorac Surg. 2008;85:909–914.

Meta-analysis; Atrial fibrillation; Surgical ablation; Cardiac surgery

© 2010 Lippincott Williams & Wilkins, Inc.