Ablation of atrial fibrillation: surgery may show the way

This editorial refers to ‘Results of the first investigator-initiated randomised clinical trial of nMARQTM, PVACTM, and thoracoscopic ablation for paroxysmal atrial fibrillation’ by C. Sugihara et al., on pages f384–391. Catheter ablation has become a common treatment option for patients with symptomatic atrial fibrillation (AF). However, a number of questions (optimal patient selection, preferred energy source, ablation strategy, etc.) and technical challenges (to achieve durable continuous and transmural lesions in a very complex anatomy) remain, despite numerous advances in the last years. It is in this context that the study by Sugihara et al.1 is relevant as it highlights the main areas of uncertainty. In this well-designed, albeit relatively small, randomized trial using meticulous monitoring with implantable loop monitors or pacemakers before and after the ablation, three alternative treatment options are compared in patients with paroxysmal AF: catheter ablation using two ‘single-shot’ devices [both circular multipolar radiofrequency (RF) energy devices] and thoracoscopic surgical ablation [incorporating pulmonary vein isolation (PVI), empiric ablation of ganglionated plexi, a box lesion at the posterior wall of the left atrium, and left atrial appendage (LAA) excision]. Therefore, a much more extensive lesion set was performed in the surgically treated patients compared with the two catheter ablation groups where only PVI was performed. The authors show better efficacy and fewer repeat ablations and antiarrhythmic drug treatment in the surgically treated group, but it is unclear whether the higher efficacy of surgery is due to more durable PVI or to the additional lesions and exclusion of the LAA, which has been implicated with improving success of catheter ablation in patients with long-standing persistent AF.2 There was, however, a much higher incidence of complications (35%) in the surgical arm, including one death and one major bleeding requiring open sternotomy and cardiopulmonary bypass. In addition to the very small number of patients in each arm and the difference in ablation strategy between the surgical and the two catheter ablation arms, another limitation of the study is that no ‘point-by-point’ ablation arm using newer ablation technology that has been shown to improve acute outcome3 was studied. However, there is some evidence for equivalent efficacy for one of the single-shot devices used in this study compared with conventional point-by-point ablation.4 So where does this leave us? (i) Even in paroxysmal AF, the results of catheter ablation definitively need to be improved. (ii) The main problem with catheter techniques still is achieving durable PVI. (iii) For persistent forms of AF, no additional benefit of substrate modification has been demonstrated when catheter techniques were used.5,6 Pathophysiological considerations, however, would suggest that in most of these patients more advanced structural changes in the left atrium are likely to occur, and, therefore, additional substrate modification should improve the results. The most likely explanation for the lack of improvement is the inability to achieve durable transmural lesions. In the Substrate and Trigger Ablation for Reduction of Atrial Fibrillation (STAR-AF) II trial, for linear ablation, bidirectional block was demonstrated in only 75% of patients at the end of the procedure,6 both limiting the efficacy and increasing the likelihood of regular atrial tachyarrhythmia. Thus, in catheter ablation of patients with persistent forms of AF (that most likely are not all PV trigger/driver dependent), the benefit of substrate modification in the left atrium in addition to PVI is untested, and it is here that the results of surgical ablation come into play and may help develop strategies for catheter ablation. Open-heart surgical ablation (both the ‘cut-and-sew’ Cox maze procedure and the RF or cryo energy variations of this procedure) has consistently shown much better results even in patients with more advanced structural changes (e.g. valvular heart disease),7 suggesting that a concept of more extensive lesions does improve the success rates once permanent transmural lesions can reliably be achieved. The gold standard for surgical AF treatment is still the Cox maze III procedure that results in close to 90% maintenance of sinus rhythm 10 years after surgery.8 Currently, the minimally invasive forms of surgical ablation—both epicardial and hybrid epicardial surgical and catheter-based endocardial ablation procedures—are still less effective, although in most studies with this approach, only PVI was performed, and have relevant safety disadvantages (e.g. conversion to sternotomy and major bleeding),9 possibly due to technical limitations and the learning curve of the operators. In addition, a decrease in the left atrial function, presumably due to the more extensive lesions, was observed even following limited surgical ablation.10 For the foreseeable future, surgical ablation will play the most relevant role in clinical practice in patients with AF undergoing surgery for other reasons (e.g. bypass surgery and valvular disease). However, due to the continuous and more durable lesions, surgery is the ideal (and at the moment the only) tool to test ablation strategies for more advanced stages of atrial disease in persistent AF, and it may lead to the development of more effective catheter ablation techniques in the future. Sugihara et al.1 are to be commended for undertaking this difficult study and for underlining that surgery may be a helpful tool for better understanding of the ablation strategies. Conflict of interest: J.C.G. is a consultant for St. Jude Medical, Medtronic, Biotronik, Biosense Webster and Boston Scientific, Boehringer Ingelheim, Daiichi Sankyo, and Pfizer and has received speaker fees from AstraZeneca, Bayer, Boehringer Ingelheim, Daiichi Sankyo, Pfizer, Sanofi Aventis, Biotronik, Boston Scientific, Medtronic, and St. Jude Medical. T.K. is a consultant for Atricure. The opinions expressed in this article are not necessarily those of the Editors of Europace or of the European Society of Cardiology.