Attempting to cure atrial fibrillation during mitral valve surgery: can we measure a benefit?

Atrial fibrillation (AF) and mitral valve (MV) disease are often intertwined in cause and consequence, but there is an increasing appreciation of the morbidity associated with the rhythm disorder itself. The headline risk of stroke resultant from AF has long been appreciated, but the reduction in cardiac output ensuing from arrhythmia may be the tipping point into heart failure. In addition, AF is also linked with early cognitive decline. Hence, there are several reasons to believe that patients who remain in sinus rhythm after mitral valve surgery may have lower cardiac morbidity. Around one-third of these patients presenting for mitral valve surgery will have concomitant atrial fibrillation, and the prospect of attempting to establish sinus rhythm by surgical ablation or Maze procedures is certainly appealing. Therefore, the paper published by Rostagno et al. in this issue of European Heart Journal—Quality of Care and Clinical Outcomes1 is of particular interest, highlighting real-world experience of this treatment strategy, which comes on the back of increasing understanding of the results of adjunctive surgical therapy for atrial fibrillation. Very high proportions of patients maintaining sinus rhythm in the long-term following surgical AF ablation have been reported.2 In a large meta-analysis by Phan et al., concomitant surgical AF ablation in patients undergoing mitral valve surgery appeared both safe and effective at restoring sinus rhythm, with overall rates of freedom from AF of 75.5 vs. 26% at 12 months.3 Indeed, the indication for initial surgery does not appear to dictate the surgical AF ablation outcomes and similar outcomes are achieved in patients undergoing other forms of cardiac surgery including coronary artery bypass grafting or aortic valve replacement.4 The Cox-Maze 3 procedure has often been referred to as the ‘gold standard’ of surgical AF ablation, but as well as its extreme technical challenge, it has been associated with frequent requirements for permanent pacing, increased bleeding risk, and longer cardiopulmonary bypass times. In 2004, Gaynor et al. used radiofrequency ablation catheters to replace the surgical incisions of the Cox-Maze 3 procedure, resulting in reduced procedure time without compromise in short-term efficacy.5 The reduction of stroke risk in the long-term, following a Cox-Maze procedure has also been demonstrated.6 With the introduction of surgical radiofrequency ablation tools, the development of the ‘Cox-Maze 4’ procedure followed. This includes an extensive lesion set: pulmonary vein isolation, left atrium (LA) floor and roof line, right atrium appendage and tricuspid valve (TV) annulus lesions, mitral valve isthmus line as well as targeting other potentially arrhythmogenic areas, e.g. the cardiac portion of the superior vena cava. When assessing which lesion sets are most effective, Gelsomino et al. showed higher success rates with inclusion of mitral isthmus line, performing biatrial ablation, and use of bipolar RF catheters.7 Recent work showed no significant difference in mortality rates among patients undergoing MV surgery in combination with RF ablation in this manner and LA appendage excision,8 but a significant quality-of-life improvement was reported and 78% of patients were free of AF at 1 year. The introduction of these radiofrequency ablation tools for cardiothoracic surgery mitigates many of the drawbacks with the Cox-Maze, but questions remain as to whether patient benefit is maintained. This was recently tested in a randomized trial comparing mitral valve surgery with or without surgical ablation therapy in patients undergoing mitral valve replacement with atrial fibrillation. Arrhythmic results were impressive, with over double the numbers of patients remaining in sinus rhythm in the ablation group compared with control. Yet this was not associated with improvements in quality of life or reductions in complications, and indeed there was a significantly higher rate of pacemaker implantation in the ablation group (21.5 vs. 8.1%).9 In another large multicentre randomized control trial, similar arrhythmic results were achieved; however, there was no significant difference in 1 year all-cause mortality between patients undergoing cardiac surgery with surgical left atrial ablation vs. without.10 Minimally invasive Cox-Maze procedures have also been developed. Moten et al. showed an even more impressive 87% sinus rhythm success rates after 1 year, with a minimally invasive cryoablation-maze procedure being performed through a comparatively small right infra-mammary incision.11 A systemic review by Hyung Gon Je et al. recently compared minimally invasive surgical treatment for AF.12 Reported freedom from AF at 1 year from the Cox-Maze was 93, vs. 80% from ‘beating heart’ epicardial ablation (where cardiopulmonary bypass is not used), vs. 70% if a hybrid surgical and endocardial procedure is performed. Despite all the technological advancements inherent in this hybrid approach, the Cox-Maze 3 remains a gold standard procedure. Thoracoscopic techniques are also provoking interest. In a recent series by Probst et al.,13 a thoracoscopic left atrial ablation technique achieved maintenance freedom from AF at 1 year in 93% of patients, approaching the success rates reported from the Cox-Maze 3 series. Of these, 12% developed left atrial tachycardias, presumably due to incomplete lesion sets providing the substrate for fixed re-entry. 86% of the 37 patients who were in longstanding persistent AF at the time of the procedure maintained sinus rhythm at 1 year. But with satisfactory results now routinely being achieved by day-case endocardial ablation procedures in the electrophysiology lab, the patient selection for whom a complex thoracic surgical procedure (although with smaller incisions than a median sternotomy) is most suitable remains unclear. The key role for surgically delivered ablation of atrial fibrillation appears to be as an adjunct to a planned primary cardiac procedure. It is in this context that the real-world data presented by Rostagno et al. can be reviewed.1 The authors present a retrospective series of 173 patients who underwent mitral valve replacement and radiofrequency surgery. Their experience of maintaining sinus rhythm was very similar to that of other published data, around two-thirds of patients remained in sinus rhythm at 6 years. Remarkably, the very thorough follow-up of this study allows insights into whether there are any other benefits in staying in sinus rhythm, and the authors provocatively suggest that there may be a survival benefit to this. The data appear to support this assertion on first inspection, patients in whom sinus rhythm was established have a markedly better outcome than those who remain in persistent fibrillation. Yet those patients who relapse to AF after a period of sinus rhythm had similar outcomes to those remaining in sinus rhythm, pointing to differences in the severity of heart disease as the underlying factor both for persistence of AF despite ablation and for increased morbidity. The authors are clear about this possibility, and offer that patients with more severe disease (evidenced by concurrent TV surgery) almost never retain sinus rhythm. Further, TV surgery was strongly associated with mortality during the course of follow-up, 40% of the deceased had undergone TV surgery (compared with only 12% of the survivors), and could be strongly associated with persistence of AF in multivariate analysis. Thus, these data suggest what we often assume: more extensive cardiac disease is associated with harder-to-treat AF. Any apparent mortality benefit from sinus rhythm maintenance could well be a reflection of the better cardiac function required to maintain sinus rhythm. The limitations of such an observational study are manifest and the lack of a prospective control group renders meaningful comparisons between groups speculative at best. Nevertheless, this study does have the key benefits of showcasing what may be achieved by ablation concomitant with mitral valve surgery. So should we now perform ablation routinely on all atrial fibrillation patients undergoing cardiothoracic surgery? Certainly this additional procedural step seems to be safe and effective, and can be performed with hope of worthwhile and sustained outcomes even outside of the ramified settings of randomized trials, and adds to the perception put forward by the guidelines. In the 2009 International Society of Minimally Invasive Cardiothoracic Surgery consensus statement, a class 1, level A recommendation put forward for surgical AF ablation with concomitant MV surgery to restore incidence of sinus rhythm in the short- and long-term follow-up period,14 but this single recommendation fails to capture the uncertainties surrounding the patient's perceptions of outcomes, which are given 2a recommendations. The EHRA/HRS/ECAS consensus statement in 2012 gave a class 2, level C indication for surgical AF ablation for symptomatic AF patients undergoing cardiac surgery.15 Similarly, the more recent AHA/ACC and HRS guidelines from 2014 state that surgical maze procedures are reasonable in selected patients undergoing cardiac surgery.16 For now, these recommendations continue to appear appropriate, but with level of evidence upgraded to B. Longer follow-up of recent randomized studies such as Gillinov et al.,8 with additional quality-of-life assessments may provide clarity of symptomatic and heart failure benefits that might be expected in patients maintaining sinus rhythm. Conflict of interest: M.