Journal of the American College of Cardiology Vol. 58, No. 7, 2011© 2011 by the American College of Cardiology Foundation ISSN 0735-1097/$36.00Published by Elsevier Inc. doi:10.1016/j.jacc.2011.05.014

EDITORIAL COMMENT

Intracranial EmboliAssociated With CatheterAblation of Atrial FibrillationHas the Silence Finally Been Broken?*

Jonathan S. Steinberg, MD, Suneet Mittal, MD

New York, New York

Catheter ablation has become an accepted treatment ap-proach for drug-refractory, symptomatic atrial fibrillation(AF). The technical aspects of the procedure have evolved,but to a remarkable degree, the objective of the procedure isnearly the same as when it was first introduced: the electricalisolation of the principal trigger sites of AF, the pulmonaryveins (PVI). The procedure has an excellent success rate andacceptable safety profile, although AF may not be perma-nently eliminated even if response is initially complete (1).

See page 681

There are, however, a number of complications that areassociated with the procedure. One of the rare but mostdreaded complications of the procedure is systemic embo-lism, which typically presents as a transient ischemic attack(TIA) or stroke. The mechanism by which stroke occurs islikely thromboembolism in the majority of instances; how-ever, embolism of air and/or coagulum likely contributes aswell (2). To minimize embolic complications, most electro-physiology laboratories have developed standards for pre-,intra-, and post-procedural anticoagulation as well as man-agement of intracardiac sheaths and catheters. Nonetheless,the most recent worldwide survey on catheter ablation ofAF reported a 0.71% incidence of TIA and 0.23% incidenceof stroke (3). To date, the incidence of thromboembolismhas not been thought to vary with the energy source used forablation.

*Editorials published in the Journal of the American College of Cardiology reflect theviews of the authors and do not necessarily represent the views of JACC or theAmerican College of Cardiology.

From the Valley Heart and Vascular Institute, Columbia University College ofPhysicians & Surgeons, New York, New York. Dr. Steinberg is a consultant forBiosense-Webster, Cameron Health, Medtronic, St. Jude Medical, Ortho-McNeilJanssen, Sanofi, and Stereotaxis, and has received research support from Biosense-Webster and Medtronic. Dr. Mittal is a consultant for Biosense-Webster, Biotronik,

Boehringer-Ingelheim, Boston Scientific, Medtronic, and St. Jude Medical), and hasreceived research support from GlaxoSmithKline and Biosense-Webster.

As the technique for AF ablation has matured, newmodalities of energy delivery are being introduced, primarilyin an effort to make this complex procedure more efficientand accessible. Two of these new entries, cryoballoon andmultielectrode phased radiofrequency (RF) pulmonary veinablation catheter (PVAC), in comparison to the standardapproach of RF delivery using an open irrigated tip catheter,are the focus of inquiry in the important study of Siklody etal. (4) published in this issue of the Journal. The data arederived from an observational series of 74 patients under-going ablation at 3 centers in Europe; the study was stoppedprematurely when an independent ethics committee noted adramatically higher rate of silent brain infarction followingablation in the PVAC group.

The patients in this study were similar to many priorpublished series, although slightly older. More than 60% ofthe patients had paroxysmal AF, which is typical, and theremainder had persistent AF. All patients underwent PVIalone without any additional linear or nonlinear left atrialablation. The groups were well matched. All patients hadsimilar pre-, intra-, and post-procedural care, includinganticoagulation, level and duration of energy delivery, andperformance of transesophageal echocardiography (TEE) toexclude pre-existing thrombus. They did not undergo in-tracardiac echocardiography. The 27 patients who under-went irrigated RF ablation had lesions placed in a circum-ferential fashion around each ipsilateral pair of PVs. The 24patients who underwent cryoballoon ablation had balloonssized according to TEE-measured diameters; touch-upusing a focal cryoablation catheter was performed if isolationwas not achieved with the balloon alone. The PVAC wasused in 24 patients. It is a circular mapping and ablationsystem capable of duty-cycled phased unipolar and bipolarRF delivery at low energy levels to specific electrodes orpairs of electrodes, without internal or external irrigation.All patients, irrespective of technique, were required toachieve complete PVI, and indeed, all met this proceduralendpoint. Ultimate clinical outcome, that is, AF suppres-sion, was not reported.

What is reported is a marked disparity in incidence ofnew embolic events as detected by brain magnetic resonanceimaging (MRI). All patients underwent a pre-ablation (1day prior) and a post-ablation (within 2 days) scan. Newembolic events were documented in 7.4% of irrigated RFpatients, 4.3% of cryoballoon patients, and 37.5% of PVACpatients, statistically different among groups (p � 0.003).The PVAC patients had between 1 and 5 new lesions(median � 3) with a wide intracranial distribution, consis-tent with an embolic source. Based on an examination by anon-neurologist physician, all patients were asymptomaticwithout overt neurological findings in the early aftermath ofablation, although sophisticated examination and testingwere not performed. The use of cryoenergy was not immuneto silent infarction, confirming prior results (5). The use of

PVAC was the only factor predictive of new embolic events.

690 Steinberg and Mittal JACC Vol. 58, No. 7, 2011Intracranial Emboli and Catheter Ablation of AF August 9, 2011:689–91

The findings in this study are startling and raise 2 veryimportant issues: 1) the safety of ablation for AF in regardto silent brain infarcts and its long-term clinical impact; and2) whether newer alternative techniques have unexpectedand potentially serious deleterious consequences.

It has only recently been recognized that an overt TIA orstroke likely represents only the “tip of the iceberg” when itcomes to intracranial emboli resulting from catheter abla-tion of AF. MRI of the brain performed within the first fewdays of ablation can demonstrate clinically silent intracranialemboli in up to 14% of patients (6,7). Although a singleembolic lesion is typically observed, some patients experi-ence multiple emboli. Against this historical context, theobservation in this study that a median of 3 new emboliclesions was observed in 37.5% of PVAC patients is evenmore alarming. The unresolved question is the clinicalsignificance of these silent brain infarcts. In the presentstudy, there was no control group, and neither repeat MRInor neuropsychological testing was performed. Nonetheless,evidence suggests that these silent infarcts may not bebenign. For example, in population-based studies, the pres-ence of silent brain infarcts has been associated with worseperformance on neuropsychological tests, a steeper declinein global cognitive function, and doubling of the risk ofdeveloping dementia (8). Patients must be informed of thisrisk and its uncertain consequences, and formal indicationsfor the ablation procedure must weigh this risk, amongothers, relative to the very real symptom benefit for patients.It is imperative that new studies focus on very compre-hensive, sensitive and reproducible neuropsychologicaltesting, performed serially over at least several monthsand perhaps longer, in order to examine if, when, and towhat extent these infarctions impact brain function.Serial MRI scans would help confirm the chronicity ofthe lesions. Only when armed with functional data canthe medical and patient communities properly make atruly informed comparative assessment of treatment mo-dalities for AF. It should be acknowledged, however, thatAF alone can cause silent cerebral embolism (9), andsuperior suppression of AF versus other treatments byablation may in fact produce a net benefit to patients, andthus controlled studies are critical. The final answersregarding consequences of these events may take manyyears to be fully understood.

Are these results so definitive that they have dealt a fatalblow to the new PVAC technology? The ethics committeeprematurely terminated the study, so clearly these membersfelt convinced that patient safety was jeopardized. However,the small sample size means that the confidence intervalsaround the point estimate of embolic risk (not reported bythe investigators) will be quite wide (true for all groups),making accurate risk assessment challenging. Thus, it iscrucial to determine whether the findings reported bySiklody et al. (4) are reproducible with different investiga-tors and centers, using larger samples. To this point, a very

recent study by Gaita et al. has just reported virtually

identical rates of silent cerebral embolism (including aseveral-fold higher rate for PVAC) when comparing thesame modalities of ablation in 108 patients with paroxysmalAF (10). Although clinically apparent acute cerebral throm-boembolic events do not appear to occur at a higher ratethan expected (11,12), it is important to take a step back andask whether the PVAC technology has incremental valueversus alternative approaches. These 2 compelling recentreports indicate that this technology may have substantiallygreater risk than competing energy delivery systems, and itwould be difficult to justify its use until greater clarificationregarding mechanism and long-term consequences of silentcerebral embolism are clarified and addressed, or the find-ings are refuted.

What potential explanation underlies a several-fold in-crease in embolic events when using the PVAC catheterversus other techniques? Intuitively, restricting ablationdeliveries to the fewest necessary to accomplish proceduralgoals is desirable (13); however, the PVAC and irrigated RFsystems had similar cumulative energy levels. Proceduretime is important because catheter dwell time in the leftatrium exposes the patients to thrombus formation onablation apparatus, air entry, and the inevitable variability indegree of anticoagulation. However, the PVAC patientshad the shortest procedures. There is a nagging concern thatreliance on duty-cycling and passive tissue cooling forprevention of thrombus formation and charring when noirrigation is applied during RF delivery may not in fact beaccomplished routinely. Although, only a minority hadvisibly detectable material, there is no assurance that char orthrombus had not formed and already embolized, and it isnot clear how diligent or meticulous was the search formaterial adherent to the catheter. The nature of the PVACsystem promotes current density similar to conventionalelectrodes and power settings (11), which in turn mayfacilitate heating concentrated in small anatomic regions orheating of blood elements if tissue contact with the elec-trode is absent. It is unknown whether the materials of thePVAC ablation catheter or the manipulation during theprocedure contributes to the risk as well.

The rates of silent embolic events for the irrigated RFcatheter and cryoballoon are also higher than desirable.Should present-day practice change in any manner to takeinto account the risk of these events? We suggest thefollowing initiatives.

1. It is a given that meticulous attention should be em-ployed with respect to anticoagulation throughout theablation process, including the sheath systems typicallyused. It may be valuable to test more intensive orcustomized (for ablation technique) anticoagulation reg-imens including higher activated clotting time targets,addition of antiplatelet medications. or use of the non-vitamin K antagonist anticoagulants as bridging or in-

traprocedure agents.

691JACC Vol. 58, No. 7, 2011 Steinberg and MittalAugust 9, 2011:689–91 Intracranial Emboli and Catheter Ablation of AF

2. Catheter and ablation system designs should be thor-oughly reassessed in the laboratory and with animalmodels to better understand by which mechanism thesecomplications may occur, so that redesign could beundertaken if warranted.

3. The routine use of intracardiac ultrasound deservesfurther scrutiny as a means of early detection of throm-bus before embolization (14).

4. Avoidance of intraprocedural cardioversion has beensuggested (7).

5. It seems premature to require that all patients undergopre- and post-ablation brain MRI, but a large-scaleeffort, probably via a multicenter registry, must beundertaken to more accurately depict the incidence,infarction pattern, and risk factors of embolic eventsindependent of technique. One also wonders whetherother left-sided atrial and ventricular ablations are asso-ciated with silent embolism as well.

6. It is possible that upstream filters may be required toprevent undesirable material from reaching the cranialcirculation.

7. Finally, the inequality of distribution of events amongthe different ablation techniques strongly suggests thattesting be part of the regulatory process before a newdevice or system reaches the market.

There is no question that AF ablation is an importanttherapeutic advance, but unintended consequences fromcomplex invasive procedures should be thoroughly investi-gated and adjudicated. Our first and foremost concern mustbe both short- and long-term patient safety.

Reprint requests and correspondence: Dr. Jonathan S. Steinberg,Columbia University College of Physicians and Surgeons, 1111Amsterdam Avenue, New York, New York 10025. E-mail:jss7@columbia.edu.

REFERENCES

1. Shah AN, Mittal S, Sichrovsky TC, et al. Long-term outcome

following successful pulmonary vein isolation: pattern and predictionof very late recurrence. J Cardiovasc Electrophys 2008;19:661–7. m

2. Asirvatham SJ. Ablation for atrial fibrillation: can we decrease throm-boembolism without increasing the risk for bleeding? Circulation2007;116:2517–9.

3. Cappato R, Calkins HM Chen SA, et al. Updated worldwide surveyon the methods, efficacy, and safety of catheter ablation for humanatrial fibrillation. Circ Arrhythm Electrophysiol 2010;3:32–8.

4. Siklody CH, Deneke T, Hocini M, et al. Incidence of asymptomaticintracranial embolic events after pulmonary vein isolation: comparisonof different atrial fibrillation ablation technologies in a multicenterstudy. J Am Coll Cardiol 2011;58:681–8.

5. Neumann T, Kuniss M, Conradi G, et al. MEDAFI-Trial (Micro-embolization during ablation of atrial fibrillation): comparison ofpulmonary vein isolation using cryoballoon technique vs. radiofre-quency energy. Europace 2011;13:37–44.

6. Lickfett L, Hackenbroch M, Lewalter T, et al. Cerebral diffusion-weighted magnetic resonance imaging: a tool to monitor the throm-bogenicity of left atrial catheter ablation. J CardiovacElectrophysiol2006;17:1–7.

7. Gaita F, Caponi D, Pianelli M, et al. Radiofrequency catheter ablationof atrial fibrillation: a cause of silent thromboembolism? Magneticresonance imaging assessment of cerebral thromboembolism in pa-tients undergoing ablation of atrial fibrillation. Circulation 2010:122:1667–73.

8. Vermeer SE, Prins ND, Heijer TD, Hofman A, Koudstaal PJ,Breteler MMB. Silent brain infarcts and the risk of dementia andcognitive decline. N Engl J Med 2003;348:1215–22.

9. Das RR, Seshadri S, Beiser AS, et al. Prevalence and correlates ofsilent cerebral infarcts in the Framingham offspring study. Stroke2008;39:2929–35.

10. Gaita F, Leclerq JF, Schumacher B, et al. Incidence of silent cerebralthromboembolic lesions after atrial fibrillation ablation may changeaccording to technology used: comparison of irrigated radiofrequency,multipolar nonirrigated catheter and cryoballoon. J Cardiovasc Elec-trophysiol 2011 Mar 31 [E-pub ahead of print].

11. Scharf C, Boersma L, Davies W, et al. Ablation of persistent atrialfibrillation using multielectrode catheters and duty-cycled radiofre-quency energy. J Am Coll Cardiol 2009;54:1450–6.

12. Beukema RP, Beukema WP, Smit JJJ, et al.. Efficacy of multi-electrode duty-cycled radiofrequency ablation for pulmonary veindisconnection in patients with paroxysmal and persistent atrial fibril-lation. Europace 2010;12:502–7.

13. Khan A, Mittal S, Kamath GS Garikipati NV, Marrero D, SteinbergJS. Pulmonary vein isolation alone in patients with persistent atrialfibrillation. J Cardiovasc Electrophysiol 2011;22:142–8.

14. Steinberg JS, Maleki K, Koneru J, Chaudhry F. Intracardiac ultra-sound. In: Calkins H, Jais P, Steinberg JS, editors. A PracticalApproach to Catheter Ablation of Atrial Fibrillation. New York, NY:Wolters Kluwer Lippincott Williams & Wilkins, 2008:57–82.

Key Words: atrial fibrillation y catheter ablation y embolic events yagnetic resonance imaging.