OUP user menu

How do we discover the unknown unknowns? Screening ‘Well’ patients after catheter ablation for atrial fibrillation

David D. Spragg, Hugh Calkins
DOI: http://dx.doi.org/10.1093/europace/eur242 1362-1363 First published online: 27 July 2011

This editorial refers to ‘Persistent iatrogenic atrial septal defect after pulmonary vein isolation by cryoballoon: an under-recognized complication’ by N-Y. Chan et al., on page 1406.

[T]here are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns – the ones we don't know we don't know. -Former US Secretary of Defense Donald Rumsfeld

Atrial fibrillation (AF) remains a therapeutic challenge for millions of patients and their treating physicians. Advances in pharmacotherapy, including the development of novel antiarrhythmics1 and anticoagulants,2 have provided clinicians newer—but not necessarily better—treatment options for AF patients. There has been a concurrent evolution in catheter ablation for AF; novel ablation techniques have come, and in some cases gone,3 over the last several years. Whether these novel techniques will translate into improved safety and efficacy of AF ablation remains to be seen.

While ablation modalities used in catheter-based therapy for AF are evolving, the physical introduction of those catheters into the left atrium (LA) remains a nearly universal constant. This is true for standard radiofrequency ablation approaches [i.e. wide, circumferential lesion sets to isolate the pulmonary veins (PV)], for more targeted LA ablation (of complex fractionated atrial electrograms, sympathetic ganglia, or putative rotors, for example), and for newer methods of achieving PV isolation (cryoablation, cyroballoon ablation, laser, microwave, robotically assisted catheter ablation, and so on). Each of these approaches relies on the delivery of one (and frequently more than one) catheter into the LA. Almost always, this is achieved through transseptal puncture and catheter placement across the atrial septum. The physiological consequences of transseptal catheter placement, and how catheter size dictates those consequences, is one of the many unknowns surrounding current AF ablation strategies.

In this issue of the Journal, Chan et al.4 address one of the known unknowns: the physiological consequence of puncturing the atrial septum during the course of AF ablation. They provide data from 13 patients who underwent PV isolation using cyroballoon ablation. All patients underwent transoesophageal echocardiography (TEE) prior to the procedure, and then both 6 and 9 months after the procedure. Single transseptal puncture was performed using a Brockenbrough needle, with introduction of an 8Fr sheath into the LA, followed by upsizing to a steerable 15Fr (outer diameter) sheath that was used for the ablation itself. The large majority of veins were isolated with the cyroballoon catheter alone (49 of 55), with an average of 2.6 freezing cycles per vein; six veins required cryoablation with an 8 mm tipped catheter for complete isolation, however. Procedures were performed off of warfarin, with heparin therapy during the procedure. Data about the use of other peri-procedural medications that could impact on LA pressure (diuretics) or tissue healing (steroids) are not provided.

The authors report that there were no instances of patent foramen ovale or atrial septal defect (ASD) in their small series of patients prior to the procedure. However, 6 months post-ablation there were five patients (38%) with a visible ASD on TEE, with a mean maximal dimension of 5.5 mm. At 9 months post-ablation, one patient had a resolution of the ASD, while four had persistent (but slightly smaller) septal defects. There were no instances of right-to-left shunting, paradoxical embolism, migraine headache, or other clinically apparent consequences of the ASDs. As such, the authors provide results that are prima facie reassuring. However, the number of patients in this report is small, with only three persistent AF patients included. There is no control group undergoing standard RF ablation with double-puncture transseptal access provided. Finally, the details surrounding the transseptal access itself (LA dwell time; location of the actual puncture site) are somewhat limited. Such details almost certainly influence whether lasting septal defects are generated.

Indeed, there is published evidence that the method of crossing the atrial septum matters in the potential development of iatrogenic ASDs following PV isolation. Double-transseptal punctures with relatively smaller sheaths (7 and 8Fr) have not been shown to cause persistent iASDs,5 whereas single punctures with larger sheaths6 or with two sheaths across a single puncture site5 can result in lasting septal defects. It is reasonable to suspect that the location of the puncture site (i.e. perforation of thin, membranous septum versus a thicker, more muscular part of the septum), the duration of puncture, and the type of sheath manipulation across the puncture site may all impact the development of lasting iASDs. Unfortunately, these data are not provided in what otherwise is a well-presented, albeit limited, description from Chan and colleagues.

The data that are provided are consistent with previously published reports suggesting that larger sheath sizes spanning the atrial septum are more likely to cause lasting septal defects. Whether these iASDs result in downstream clinical consequences—right-to-left shunting, paradoxical embolism, or left-to-right shunting to a degree that remodels pulmonary vasculature—will require more systematic analysis or larger patient series. The authors acknowledge this. At present, the data suggest that practitioners can continue current practices (including cyroballoon ablation) without fearing predictable injury from the iASDs that likely result from large-sheath or multiple-sheath techniques.

Donald Rumsfeld, in his characterization of US intelligence about Iraq's putative weapons of mass destruction in 2002, unwittingly provided a construct that aptly describes research in AF ablation (and indeed, any health-sciences investigation). There are aspects of AF ablation that are known. There are aspects that are the subject of active investigation (the known unknowns). Finally, there are the unknown unknowns—those questions surrounding AF ablation that have yet to be asked, or even considered. Despite its limitations, the current submission from Chan and colleagues, by thoroughly examining patients who have apparently tolerated the procedure well, is the type of study that may fortuitously uncover some of the unknown unknowns surrounding AF ablation. There are an increasing number of similarly oriented studies that have been recently published: microembolic showering of the central nervous system,7,8 mediastinal and peri-oesophageal injury,9 narrowing of the PVs,10 and iASDs5,6 all have been observed in patients who apparently tolerated the procedure well and who were free of clinical sequellae. It may be through surveys of apparently ‘well’ patients following AF ablation, in which non-invasive tools are used to discover unintended consequences stemming from the procedure, that some of the unknown unknowns are brought to light. It is the unknown unknowns of AF ablation, of course, that should concern us the most.

A final issue to consider is how we, as an electrophysiology community, should move forward in investigating the clinical significance of iASDs. First, although we do not feel that the results of this small study should alter clinical practice or the use of the cryoballoon, we feel it is important for electrophysiologists to have this potential complication in mind as they care for patients. If patients present with complications that could be attributed to an iASD, they should be encouraged to publish these case reports or case series. Second, as the authors note increased vigilance with screening echocardiograms is appropriate for patients who have undergone a cryoballoon ablation procedure or an ablation procedure in which the two catheter-one transeptal puncture approach is used and the likelihood of an iASD is increased. Third, this remains an area of need of continued clinical study for those with the energy and interest to carry out clinical research on this topic. And finally, the results of this study should encourage manufacturers of ablation tools to continue their efforts of minimize the size of the sheaths and catheters required to perform catheter ablation of AF.

Conflict of interest: H.C. reports receiving honoraria relevant to this topic and serving as a consultant to Medtronic and Biosense Webster. There are no other relevant conflicts of interest for other authors.


  •  The opinions expressed in this article are not necessarily those of the Editors of Europace or of the European Society of Cardiology.


View Abstract