Thoracoscopic maze: yes, we can, but should we?

This editorial refers to ‘Thoracoscopic ablation for the treatment of atrial fibrillation: a systematic outcome analysis of a multicenter cohort’, by C. van Laar et al., on pages 893--899. Catheter ablation (CA) has become the standard therapy for rhythm control in atrial fibrillation (AF), as it has proven more effective than antiarrhythmic drugs, with a relatively low complication rate. Pulmonary vein isolation (PVI) is the cornerstone of the procedure, particularly in patients with paroxysmal AF (PAF) and early persistent AF (PersAF). However, achieving long-term successful PVI remains a challenge, and the ablation endpoint in long-standing persistent AF (LSPAF) continues to be controversial. Given these limitations, the good outcomes achieved by open-chest Cox-Maze surgery fuelled the search for a ‘minimally’ invasive surgical alternative to CA that could achieve transmural and permanent lesions to result in a better success rate. Small randomized studies1–3 and a non-randomized cohort4 comparing minimally invasive surgical ablation (SA) with CA outcomes have suggested higher efficacy of SA, but at the price of increased complications. The European Society of Cardiology/European Association for Cardio-Thoracic Surgery (ESC/EACTS) guidelines for AF5 consider SA a Class IIa indication to achieve improvement in symptoms at the same level as CA in patients with PersAF or LSPAF, or in CA failure when refractory to antiarrhythmic drug therapy. However, the reported SA experience is very limited, and the results and complications vary widely from one centre to another. Due to this limited availability of data, the study presented by van Laar et al.6 in this issue of EP Europace contributes much needed information. The authors report the results of totally thoracoscopic maze (TTM) SA in a cohort of 475 patients with AF recruited in three hospitals between 2012 and 2015, the largest multicentre series to date with a homogeneous surgical technique. Similar results were reported at all three centres. TTM consisted in bilateral epicardial PVI with bipolar radiofrequency clamp. A posterior box and exclusion of left atrial appendage was routinely performed. A bicaval lesion or a lesion to the mitral valve (trigone lesion) was performed, depending on surgeon preference. The left atrial appendage was excluded using a stapler device or a clip. In their series, 52% of patients had PAF, 36% PersAF, and only 12% had LSPAF; nearly half (49%) had previously undergone a failed CA. Complications included a need for conversion to sternotomy due to bleeding (1.7%); major complications (2.5%) were hemothorax, tamponade, empyema, and stroke. Minor complications (5.5%) included pleural effusion and pacemaker implantation. The success rate observed in this study is comparable to that reported in other series,1,7 despite a wide variation between cohorts in the type of AF and the proportion of redo cases included (nearly half of this cohort), as well as the quality of the follow-up and scrutiny for recurrences. On the other hand, the complication rate is in the lower range reported by other studies.1,3,7,8 In summary, the information provided confirms that TTM may obtain better results than a single CA procedure, but with a higher complication rate and a still-significant rate of recurrence, particularly in LSPAF. Achieving transmural lesions is challenging even in surgical series, due to the presence of epicardial fat and the need for unipolar radiofrequency applications to apply the lines at the posterior box. To overcome these limitations, simultaneous or sequential TTM plus CA, the so-called ‘hybrid procedures’, have been proposed. However, there are no conclusive data about the benefits of these approaches. Indeed, a recent meta-analysis suggests a comparable efficacy but with higher complication rates.9 How do the results of TTM compare with the current standard, the CA approach? In theory, TTM can offer more transmurality of the lesions, achieving better long-term results and a lower risk of long-term cerebrovascular events due to appendage exclusion, at the price of a more complex procedure (associated with higher morbidity). However, these assumptions remain to be proven in large multicentre studies, which will be very difficult to conduct. For example, the FAST II study10 was prematurely terminated after enrolling 21 of the 90 patients initially planned. On the other hand, a recent randomized study11 challenged the concept of greater efficacy with SA. Although the study was not powered to detect superiority, a trend was observed towards higher efficacy in the CA arm (59% vs. 29% in the SA arm) at 2-year follow-up and a higher SA complication rate (21% vs. 0% in the CA arm). Continuous monitoring with an implantable recording device to better detect AF recurrences may explain the unexpectedly low SA success rate reported in their cohort. Another challenge in comparing SA and CA is related to the evolution of CA techniques and approaches. In recent years, recurrences after CA seem to have decreased significantly compared with historical series, likely as a result of major changes in technology. The development of contact force measurement ensures better catheter contact and may result in more transmural lesions.12 Single-shot techniques, particularly cryoballoon ablation, are being widely applied with good results, and personalized approaches targeting areas of fibrosis or rotor ablation are being tested. What is already known is that CA ablation is less invasive and results in lower morbidity and a shorter hospital stay.3 Considering the need for re-ablation, TTM cost-effectiveness in the long run when compared with CA has not yet been established. Finally, compared with the volume of published CA results, the reported SA data come from relatively small series recruited in highly experienced centres; those results may be not reproducible at large scale. Anyone familiar with such procedures requiring great surgical skill and expertise can understand that things can go very wrong in less experienced hands. In contrast, a huge number of centres are performing CA to treat AF, and the results of large registries show consistent results and a relatively low complication rate. Compared with a surgical procedure, a single-shot cryoballoon procedure, for example, is much less challenging, although the potential for complications is still present. Another important aspect, from the health care system point of view, is the high prevalence of AF in an ageing population. Dealing with the AF epidemic that is expected in coming years poses a significant challenge to health care systems worldwide, and CA seems to offer a technique that is simpler than surgery, can be applied by an electrophysiologist with average experience, and permits the treatment of large numbers of patients with reproducible results at acceptable cost and patient safety. Cardiac surgeons must deal with many other procedures and the capability to focus on treating AF is limited. The need for greater surgical expertise and surgeons devoted to AF therapy, and the costs and sheer number of centres involved, could make SA unachievable on a large scale. In summary, TTM SA is a solid therapeutic option with reasonable results in improving patient symptoms and decreasing recurrences but is probably not the first option in most AF patients. Whether the use of TTM is justified in patients with very low chances of CA success remains to be tested in randomized studies. While waiting for better data, it would likely be wise to perform a first attempt with CA and reserve SA as a second option. This does not mean that experienced centres, with good surgical results, should not consider SA as a first option in selected patients. However, this centre-specific approach should not yet be generalized. Therefore, the decision to refer a patient to TTM will continue to be based on individualized analysis of the case by the physician or the AF team and patient preference after careful consideration of all potential benefits and risks. Conflict of interest: L.M. reports personal fees from Abbot Medical, Medtronic, Boston Scientific, Biotronic, Biosense Webster, outside the submitted work. The opinions expressed in this article are not necessarily those of the Editors of Europace or of the European Society of Cardiology.