CLINICAL RESEARCH
Europace (2012) 14, 653–660
doi:10.1093/europace/eus048
Ablation for Atrial Fibrillation
Acute electrical isolation is a necessary but
insufficient endpoint for achieving durable PV
isolation: the importance of closing the visual gap
Marc A. Miller, Andre d’Avila, Srinivas R. Dukkipati, Jacob S. Koruth,
Juan Viles-Gonzalez, Craig Napolitano, Charles Eggert, Avi Fischer,
Joseph A. Gomes, and Vivek Y. Reddy *
Helmsley Electrophysiology Center, Mount Sinai School of Medicine, One Gustave L. Levy Place, Box 1030, New York, NY 10029, USA
Received 16 January 2012; accepted after revision 17 February 2012; online publish-ahead-of-print 14 March 2012
Aims
Temporary, ablation-mediated effects such as oedema may cause reversible pulmonary vein (PV) isolation. To investigate this, point-by-point circumferential ablation was performed to achieve acute electrical PV isolation with an incomplete circumferential ablation line. Then, the impact of this intentional ‘visual gap’ (ViG) on the conduction
properties of the ablation lesion set was assessed with adenosine and pacing manoeuvres.
.....................................................................................................................................................................................
Methods
Twenty-eight patients undergoing ablation for paroxysmal (n ¼ 20) or persistent atrial fibrillation (n ¼ 8) were
and results
included. Pulmonary vein (PV) ablation was performed around ipsilateral vein pairs. Once acute isolation was
achieved, ablation was halted and the presence and size of the ViG were calculated. The ViG electrophysiological
properties were tested with pace capture along the ViG at 10 mA/2 ms, and assessment for dormant PV conduction
with adenosine. Despite electrical isolation, a ViG was present in 75% (n ¼ 42/56) of vein pairs (21 of 28 left PVs and
21 of 28 right PVs). There was no difference in the ViG size between the left and right PVs (22.1 + 14.2 and
17.3 + 11.3 mm, P . 0.05). Dormant PV connections were revealed by adenosine in more than a quarter
(n ¼ 12/42) of acutely isolated PV pairs, of which the majority were dependent on conduction through the ViG.
.....................................................................................................................................................................................
Conclusions
Electrical PV isolation can usually be achieved without complete circumferential ablation. However, more than a
quarter of these ‘isolated’ PVs exhibit dormant conduction—predominantly via the un-ablated ‘ViGs’ in the ablation
lesion set. These findings support the hypothesis that reversible tissue injury contributes to PV isolation that may be
acute but not necessarily durable.
----------------------------------------------------------------------------------------------------------------------------------------------------------Keywords
Atrial fibrillation † Catheter ablation † Dormant conduction
Introduction
Durable pulmonary vein (PV) isolation is the basis of catheter ablation of atrial fibrillation (AF). Nevertheless, resumption of PV-left
atrial conduction (PV reconnection) is exceedingly common and
thought to be responsible for the vast majority of post-ablation
atrial tachyarrhythmia recurrences.1 This is interesting as acute
pulmonary vein isolation (PVI) is frequently and easily obtained in
most electrophysiology labs during ablation procedure for AF.
However, in the absence of cellular death, reversible tissue
oedema or thermal stunning may account for the striking discordance between the incidence of acute (.95%) and chronic
(,30%) PV isolation.2 – 4 That is, injured, but still viable, myocardial
tissue may eventually recover its conduction properties and result
in late PV reconnection.5 Indeed, pre-clinical pathologic and ultrasound data suggest that radiofrequency (RF) ablation-mediated
cardiomyocyte and interstitial oedema occur almost immediately
after energy delivery, and extends beyond the point of ablation.6
Although the mechanism through which oedema may block cellular
conduction is not fully understood, these data suggest that
temporary RF-mediated effects, such as oedema or thermal stunning, could be responsible for reversible PV ‘isolation’. In other
words, PV conduction would resume as peri-RF application’s
oedema resolves.
* Corresponding author. Tel: +1 212 241 7114; fax: +1 646 537 9691, Email: [email protected]
Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2012. For permissions please email: [email protected].
654
We hypothesized that during circumferential PV ablation, atrial
wall oedema induced by RF applications could reversibly contribute to acute electrical PV isolation. Accordingly, we performed circumferential ablation with contiguous point-by-point lesions to
determine whether acute electrical PV isolation could be achieved
prior to the placement of a fully circumferential lesion set—that is,
leaving an un-ablated ‘visual gap’ (ViG) in an otherwise contiguous
circumferential line. Then, a combination of pacing manoeuvres
and bolus infusions of intravenous adenosine was used to
assess for potential PV reconnection through these electrically
inactive ViGs. The concomitant presence of a large ViG and
PV-reconnection would favour the role of RF-induced oedema
as the mechanism for acute and therefore reversible PVI.
Methods
All procedures were performed after obtaining written informed
consent according to the institutional guidelines at the Mount Sinai
Medical Center, New York. This study was approved by our institutional review board.
Electrophysiology study
All procedures were performed under general anaesthesia with
lumenal oesophageal temperature monitoring. Three-dimensional
(3-D) electroanatomic high-density mapping (NavX Velocity, St Jude
Medical) was performed using a 20-pole double-spiral circular
mapping catheter (Reflexion Spiral, St Jude Medical). A quadripolar
catheter was placed in the non-coronary aortic valve cusp to serve
as the location reference for the electroanatomical mapping system,
and a multipolar catheter was placed within the coronary sinus (CS).
Double transseptal puncture access was guided by intracardiac echocardiography and successfully achieved in all patients. Prior to ablation,
PV electrograms (EGMs) were recorded in each vein with the circular
mapping catheter.
Contiguous line and visual gap during
catheter ablation
Point-by-point radiofrequency ablation was performed circumferentially around each ipsilateral vein pair, 1 cm from each PV ostium with a
3.5 mm externally irrigated radiofrequency ablation catheter (Celsius
Thermocool, Biosense Webster) and a deflectable sheath (Agilis, St
Jude Medical).
Importantly, careful attention was paid to ensure that each radiofrequency lesion was intentionally placed contiguous with a prior lesion.
In general, the energy settings for the anterior left atrium were 30 –
35 W/438C (for up to 1 min) and for the posterior left atrium settings
of 20 – 25 W/438C (for up to 30 s, unless an oesophageal temperature
of 38.58C was reached). The quality of each ablation lesion was optimized by titrating the delivered energy to achieve an acute impedance
fall of 10% and a reduction of the peak-to-peak atrial EGM amplitude
of 90%. If acute PV entrance block was achieved with a ViG present,
the ablation was temporarily halted, and a 5 min waiting period ensued.
The un-ablated width of the ViG was calculated with 3-D image surface
reconstruction. If PV entrance conduction resumed during the waiting
period, the ablation was continued from the last ablation lesion until
entrance block was re-achieved.
In all cases, the left PVs were targeted for ablation prior to the right
PVs. Furthermore, although there was no strict protocol for where to
begin the contiguous ablation line, the anterior ridge of LPVs and the
M.A. Miller et al.
mid-anterior wall of right pulmonary vein (RPVs), were typically targeted as the first points of ablation.
Electrophysiological testing of the visual gap
Entrance and exit block
If the patient was in sinus rhythm, and PV entrance block (with a ViG
present) did not resume during the waiting period (5 min after PVI is
achieved), the following three manoeuvres were performed to assess
PV conduction block: (i) testing for exit block—by demonstration of
failure to conduct to the left atrium during bipolar pacing (10 mA,
2 ms) from the ablation catheter around the PV ostium, while accompanied by EGM evidence of local PV capture; (ii) testing for pace capture—
myocardial tissue capture during bipolar pacing (10 mA, 2 ms) of the
un-ablated ViG; and (iii) testing for dormant PV conduction—by assessing
for PV reconnection during the infusion of 18 mg boluses of adenosine.
If the adenosine-mediated PV reconnection was transient, repetitive
doses were administered to identify the area of breakthrough. Only
adenosine injection was performed if a patient was in atrial fibrillation/flutter after PVI.
When testing for PV exit block if local PV capture could not be accurately confirmed (i.e. local EGM proof of intravein capture), then PV
exit block was deemed ‘undetermined’. Once the manoeuvers were
performed, full circumferential PV ablation was completed in all
patients; that is, additional ablation energy was delivered to ‘close’
the ViG.
Visual gap-dependent pulmonary vein reconnection
Visual gap-dependent PV reconnection was defined when (i)
adenosine-mediated reconnection (dormant conduction) occurred
with an activation pattern of the breakthrough on the circular
mapping catheter consistent with the location of the ViG and (ii) no
adenosine-mediated reconnection (dormant conduction) recurred
once the ViG was closed.
If the dormant PV conduction persisted despite complete circumferential ablation, EGM-guided focal ablation (targeting the site of earliest
activation using the circular mapping catheter) was performed to
achieve isolation. If the area of reconnection was outside the ViG, it
was tagged accordingly and deemed non-ViG-dependent PV
reconnection.
At the end of the procedure, acute electrical isolation was achieved
in all PVs and all PVs received complete circumferential ablation (i.e.
the intentional ViGs were closed). Persistent AF patients received additional left atrial, right atrial, and CS ablation as needed. In all cases, the
left PVs were targeted for ablation prior to the right PVs. Furthermore,
although there was no strict protocol for where to begin the contiguous ablation line, the anterior ridge of LPVs and the mid-anterior wall
of RPVs were typically targeted as the first points of ablation.
Statistical analysis
Normally distributed continuous variables are presented as mean +
SD. Categorical variables are expressed as number and percentage
of patients and PVs, respectively. Differences in categorical data
were tested for statistical significance using the x 2 test.
Results
Baseline demographics
Twenty-eight consecutive patients (male, n ¼ 19) undergoing
circumferential PV ablation for symptomatic drug-refractory paroxysmal (pAF, n ¼ 20) or persistent (perAF, n ¼ 8) AF were
655
Acute electrical isolation for PV isolation
included (Table 1). At the time of the procedure, 50% (n ¼ 14) of
the patients were on a class III anti-arrhythmic drug and 57% (n ¼
16) were on either a beta-blocker or calcium channel blocker. The
prevalence of hypertension, diabetes, and obstructive coronary
artery disease was 57% (n ¼ 16), 18% (n ¼ 5), and 11% (n ¼ 3), respectively. The mean left atrial size was 22 + 4 cm2 (parasternal
long axis) and the mean left ventricular (LV) ejection fraction
was 57 + 10.8%.
Characteristics of visual gaps
Entrance block with a ViG present was observed in 75% (n ¼ 42/
56) of vein pairs (21 of 28 of left PVs and 21 of 28 of right PVs)
Table 1 Clinical characteristics
Baseline demographics
All patients (n 5 28)
Age (year)
Male
64.5 + 10.1
19 (68%)
................................................................................
Paroxysmal AF
20 (71%)
Persistent AF
Hypertension
8 (29%)
16 (57%)
Diabetes mellitus
5 (18%)
Obstructive CAD
LV ejection fraction (%)
3 (11%)
57 + 10.8
Left atrial size (cm2, parasternal)
22 + 4
Class III AAD
BB/CCB
14 (50%)
16 (57%)
Values are expressed as n (%) for categorical variables, and mean + SD for
continuous variables. AF, atrial fibrillation; CAD, coronary artery disease; LV, left
ventricular; AAD, anti-arrhythmic drug; BB/CCB, beta blocker and/or calcium
channel blocker.
(Table 2). Of those PV lesion sets that could be assessed, 92%
(n ¼ 24/26) had both entrance and exit block. The remaining
PV pairs with entrance block and a ViG were not assessed
because either AF/atrial flutter precluded pacing to assess for
exit block or local capture could not be definitively confirmed.
There was no significant difference in the size of the ViG
between the left PVs (22.1 + 14.2 mm) and right PVs (17.3 +
11.3 mm, P . 0.05). The ViGs were predominantly located
along the superior and posterior aspects of the isolating lesion
sets (Figure 1).
During the 5 min waiting period prior to pacing manoeuvres or
adenosine infusion, three PV pairs spontaneously reconnected—all
through the ViG. In all three cases, the integrity of the contiguous
ablation line along one of the edges of the ViG resulted in
re-isolation with a significant portion of the ViG still present. An
example of such an acute spontaneous PV reconnection during
the waiting period is shown in Figure 2. This example was interesting in that when the left PVs spontaneously reconnected during the
waiting period, re-isolation was achieved by placing a single ablation
lesion at the edge of the ViG opposite the prior ablation lesion that
had achieved transient isolation.
Pace capture along the ViG was present in majority of tested
PVs (68%, n ¼ 21/31). Of the 10 ViGs for which pace capture
was not present, the majority (n ¼ 7/10) occurred in PVs which
subsequently showed no evidence of dormant conduction with adenosine (see paragraph below), and one-third occurred in PVs with
dormant conduction (n ¼ 3/10).
After the waiting period, adenosine was infused; its electrophysiological impact was confirmed by observing transient atrioventricular conduction block in all cases. Twelve of the 42 of the
isolated PV pairs demonstrated adenosine-mediated PV reconnection (dormant conduction), of which two-thirds (n ¼ 8/12) were
through the ViG (Figure 3). The remaining 4 of the 12 (30%) reconnections were located at ablated segments outside the ViG and
deemed non-ViG-dependent. Figure 4 is a flow diagram of the
results.
Table 2 Electrophysiology study and ablation
................................................................................
First procedure (n ¼ 28)
PV pairs with a ViG and entrance block
42/56 (75%)
LPV ViG size (mm)
22.1 + 14.2*
RPV ViG size (mm)
Dormant conduction
17.3 + 11.3*
12 (29%)
Conduction through ViG
8 (67%)
Conduction through prior ablation
Pace capture present along tested ViGs
4 (33%)
21/31 (68%)
Re-mapping of PVs (n ¼ 6 patients, 12 PV pairs)
Total duration of follow-up (days)
Time between first and second procedure (days)
389 + 63
163 + 133
Durable PV isolation (PV pairs)
11/12 (92%)
Values are expressed as n (%) for categorical variables, and mean + SD for
continuous variables. LPV, left-sided pulmonary veins; RPV, right-sided pulmonary
veins; PV, pulmonary vein; ViG, visual gap.
*¼P value .0.05 between LPV and RPV ViG size.
Figure 1 Locations of visual gaps. Six-segment model of pulmonary vein antrum demonstrating the locations of the visual
gaps. The number within the star represents the number of
visual gaps within that location. For visual gaps that encompassed
two segments, the number is shown at the line intersecting those
segments.
656
M.A. Miller et al.
657
Acute electrical isolation for PV isolation
Follow-up and re-mapping procedure
At 389 + 63 days of follow-up, 6 of the 28 patients underwent a
second procedure with the possibility for repeat invasive mapping
of the PVs: for either clinical recurrences of atrial tachycardia/AF
(n ¼ 4), LV tachycardia ablation using a transseptal approach
(n ¼ 1) or left atrial appendage occlusion (n ¼ 1). Of the 12
vein pairs studied, 92% (n ¼ 11/12) revealed durable PV isolation.
The single PV reconnection occurred in a right PV lesion set (superior and anterior), in a patient with persistent AF who developed a recurrence of symptomatic AF. Of note, during the index
procedure, the affected right PVs were not isolated with a ViG;
that is, electrical isolation had been achieved only after placing a
fully circumferential lesion set.
Discussion
Currently, there are two basic strategies of targeting the PVs with
standard radiofrequency ablation catheters: (i) continuous circumferential ablation of the PVs and (ii) segmental PV ablation, targeting the earliest activation along a circular mapping catheter.
Although it has been accepted by some that ablation could be
halted once the requirement of acute electrical isolation was
achieved with either approach—incomplete circumferential ablation inherently leaves potential conduction gaps in the PV-left
atrial junction.7 Even small gaps in ablation lines could predispose
to clinical recurrences, since pre-clinical studies have demonstrated
that AF can propagate through even through gaps as small as
1.1 mm, and the gaps themselves can be pro-arrhythmic.8,9
Indeed, in patients who develop atrial tachycardia following circumferential pulmonary vein ablation for AF, the majority of left
atrial re-entrant atrial tachycardias are gap related.10,11
Acute vs. long-term PVI rates
Recent data suggest that acute PV electrical isolation does not
predict long-term PV disconnection. Up to half of the PVs demonstrate reconnection within 60 min of acute electrical isolation4 and
in those few studies that have assessed the durability of PV isolation after radiofrequency ablation, only a minority of veins
remain isolated.12 In both short- (1 year) and long-term (2– 5
years) clinical studies of outcomes after radiofrequency ablation,
PV reconnection rates in patients undergoing second procedures
typically .75%, and seem directly related to clinical recurrences.1,13 Many investigators have postulated that the reason for
the lack of durability in PV isolation is reversibility of the ablation
lesion set, perhaps from acute tissue oedema.
Present findings
The present study demonstrates that with careful attention to
lesion quality and contiguity during circumferential PV ablation,
(i) electrical PV isolation can typically be achieved despite an incomplete circumferential lesion set, (ii) the ViGs are, on average,
2 cm wide, and (iii) when dormant conduction is provoked by
adenosine infusion, electrical conduction typically proceeds
through the ViGs. These data suggest that acute electrical isolation
is a necessary, but insufficient, endpoint for achieving durable PV
isolation.
Electrophysiological characteristics
of visual gaps
Two interesting phenomena were observed in the un-ablated
segment of tissue inside ViGs: (i) resumption of PV–LA connection
thru the ViGs after a bolus of intravenous adenosine and (ii) failure
to pace capture the ViGs.
Based on the estimate that the perimeter of each set of circumferential lesions is 100 mm, acute electrical PV isolation was
achieved with nearly one-fifth of ipsilateral PV perimeters still
un-ablated.14 However, more than a quarter of these lesion sets
demonstrated dormant conduction as assessed by adenosine infusion—the majority through the intentional gap. Resumption of PV
conduction after adenosine (which is known to induce hyperpolarization of the resting membrane potential) is a clear indication of
the presence of viable, not fully ablated tissue in the ViGs area.
The mechanism of transient conduction block is not fully understood but it could be explained by the occurrence of RF-application-induced oedema in that area.
The utilization of pace-capture testing to identify potential conduction gaps is based on the principle that high pacing thresholds
may correlate with more effective ablation lesions.15,16 Nevertheless, a small proportion of PV pairs (5%) can exhibit persistent PV
conduction despite loss of pace capture; a phenomenon that may
be due to oedema-mediated elevations in pacing thresholds (inexcitable tissue).17,18 This theory is bolstered by our demonstration
that nearly one-third of the PVs with a ViG exhibited loss of
pace capture within the un-ablated segment of tissue.
These data are all consistent with the hypothesis that extension
of oedema beyond the ablation lesion may result in electrical iso-
Figure 2 Voltage map of spontaneous reconnection during waiting period and re-isolation with a visual gap. Electroanatomical maps in the
posteroanterior view are shown on the left and accompanying surface electrocardiogram leads (I, aVF, and V1), intracardiac electrograms
recorded from a circular multi-electrode catheter placed inside the left common pulmonary vein (SPds – SPpx) and coronary sinus from proximal to distal (CSpx– CSds) are shown to the right of each map. Left: From top to bottom, demonstrating: (i) baseline (pre-ablation) voltage map
and pulmonary vein potentials, (ii) transient pulmonary vein entrance block with spontaneous reconnection in the presence of a visual gap—
note that the activation sequence following spontaneous reconnection is the same pattern of activation that was present before the original
entrance block occurred, and (iii) pulmonary vein voltage map following spontaneous reconnection. Right: From top to bottom, demonstrating:
(i) location of the single lesion which re-isolated the pulmonary vein (opposite the prior lesion), (ii) location of pace capture testing site, demonstration of entrance block and lack of dormant conduction with a visual gap present, and (iii) final ablation lesion set and accompanying
voltage map (complete circumferential ablation). Not shown is the spiral mapping catheter in the pulmonary vein, which demonstrated that
the site of earliest activation corresponded to the location of the visual gap.
658
M.A. Miller et al.
Figure 3 Dormant conduction through the visual gap. Surface electrocardiogram leads (I, aVF, and V1), and intracardiac electrograms
recorded from a circular multi-electrode catheter placed inside the left common pulmonary vein (SPds– SPpx) and the coronary sinus from
proximal to distal (CSpx– CSds). Top and bottom panels, from left to right demonstrating: baseline left common pulmonary vein electrogamss,
pulmonary vein entrance delay, pulmonary vein entrance block, pulmonary vein exit block (intra-pulmonary vein pacing), complete AV block
with adenosine-mediated pulmonary vein reconnection (transient) and finally lack of dormant conduction once the visual gap was completed
(and re-bolused with adenosine). Middle panels, accompanying electroanatomical map in the left anterior oblique and posteroanterior views of
the left atrium with a visual gap present and following completion of circumferential ablation.
lation prior to the completion of circumferential ablation; a temporary phenomenon with a predilection for electrical reconnection.19 Thus, the ViG area must be ablated to prevent future PV
reconnection despite the presence of acute PV disconnection.
Clinical implications
Although electrical PV isolation is generally considered to be an
appropriate endpoint during AF ablation, the data from this manuscript suggest that acute electrical isolation should only be viewed
as one element of the procedural endpoint. The ability of the
intended ViG to support conduction when provoked with adenosine emphasizes the importance of avoiding gaps in ablation lines
and the need to ablate PV contiguous and circumferentially. Of
course, this recommendation would be bolstered by additional
data on the propensity for chronic reconnection of ViGs. Although
we have not conducted a series of cases in which the ViGs were
left un-ablated, it is interesting to consider the anecdotal case
shown in Figure 5 (this patient is not part of the original cohort
studied in this manuscript). After isolating the right PVs with a
ViG, and adenosine was demonstrated not to cause acute reconnection, the ViG was left un-ablated because of the proximity
of the oesophagus at this location. However, after an initial period
of symptomatic quiescence (4 months), the patient underwent a
second procedure for AF recurrences. And importantly, the only
electrical reconnection was at the right PVs at the location of
the previously un-ablated ViG.
Our data lend clinical credence to the concept that each ablation lesion should be viewed as resulting in a zone of necrosis,
but surrounded by a zone of oedema. And this oedematous
tissue can contribute to the acute isolation of the PVs—albeit reversibly. Thus, a series of discontinuous ablation lesions arrayed in
a segmental circular fashion allows for the possibility of multiple
points of acutely non-conducting, but still viable myocardium—
the substrate for chronic reconnection.
Limitations
This study has several limitations. Although we assumed that the
incidence of dormant conduction and PV reconnection would be
Acute electrical isolation for PV isolation
659
Figure 4 Flow diagram of pacing manoeuvres and adenosine.
Figure 5 An anecdotal case (not part of this original series) of chronic pulmonary vein reconnection due to a previously un-ablated visual gap.
Top panels, three-dimensional CT surface reconstruction (A) and electroanatomical map (B) in the posteroanterior view of the index pulmonary vein ablation and the re-mapping procedure, respectively. Bottom panels, surface electrocardiogram leads (I, aVF, and V1), and intracardiac
electrograms recorded from a circular multi-electrode catheter placed inside the right inferior pulmonary vein (SPds– SPpx) and the coronary
sinus from proximal to distal (CSpx– CSds). Anecdotal case (not part of this series of patients) demonstrating that during the first procedure,
the right inferior pulmonary veins were isolated with a visual gap (adenosine bolus demonstrated no evidence of dormant conduction), but the
visual gap was left un-ablated because of the proximity of the oesophagus at this location. After an initial period of symptomatic quiescence (4
months), the patient underwent a second procedure for atrial fibrillation recurrences. And importantly, the only electrical reconnection was at
the right inferior pulmonary veins at the location of the previously un-ablated visual gap. Ablation along the visual gap (red dots, (B)) re-isolated
the vein and the patient has had no atrial fibrillation recurrences in the 6 months since.
660
higher if the waiting time was prolonged and the un-ablated segments were not completed, all patients underwent completion
of circumferential ablation and closure of the ViGs. Furthermore,
although adenosine-mediated PV reconnection (dormant conduction) appears to predict clinical recurrences and chronic PV reconnection, there are currently no randomized data to support this.3
Since dormant conduction was only assessed with the infusion of
18 mg boluses of adenosine, there is the possibility that combination of adenosine and isoproterenol would have yielded different
results. However, there is conflicting data as to whether there is
incremental benefit of adding isoproterenol to adenosine when
testing for dormant conduction.20 Finally, our hypothesis of the importance of a circumferentially, contiguous lesion set to the durability of PV isolation, would have been strengthened by chronic
PV reconnection data. However, only six patients developed clinical indications during follow-up to allow for PV remapping—
though it is reassuring that of the 12 PV pairs re-tested, durable
PV isolation was present in 11 (92%).
Conclusions
A majority of pulmonary veins can be acutely electrically isolated
despite the presence of un-ablated ViGs nearly 2 cm wide. Nevertheless, more than a quarter of ‘isolated’ PV pairs exhibited
dormant conduction; predominantly through these un-ablated
gaps. These findings support the hypothesis that reversible tissue
oedema or thermal injury contribute significantly to acute PV isolation in the absence of durable cellular damage.
Conflict of interest: A.d.A. and V.Y.R. have received consulting
fees and grant support from St. Jude Medical, the manufacturer
of the electroanatomic mapping system, and Biosense Webster
the manufacturer of the ablation catheter used in this series. C.E.
is an employee of St. Jude Medical. The remaining authors report
no conflicts of interest.
References
1. Verma A, Kilicaslan F, Pisano E, Marrouche NF, Fanelli R, Brachmann J et al. Response of atrial fibrillation to pulmonary vein antrum isolation is directly
related to resumption and delay of pulmonary vein conduction. Circulation
2005;112:627 –35.
2. Medi C, Sparks PB, Morton JB, Kistler PM, Halloran K, Rosso R et al. Pulmonary
vein antral isolation for paroxysmal atrial fibrillation: results from long-term
follow-up. J Cardiovasc Electrophysiol 2011;22:137 –41.
M.A. Miller et al.
3. Datino T, Macle L, Qi XY, Maguy A, Comtois P, Chartier D et al. Mechanisms by
which adenosine restores conduction in dormant canine pulmonary veins. Circulation 2010;121:963 –72.
4. Lemola K, Hall B, Cheung P, Good E, Han J, Tamirisa K. Clinical predictors and
outcomes associated with acute return of pulmonary vein conduction during pulmonary vein isolation for treatment of atrial fibrillation. Heart Rhythm 2006;3:
1024 –8.
5. Lemola K, Hall B, Cheung P, Good E, Han J, Tamirisa K et al. Mechanisms of recurrent atrial fibrillation after pulmonary vein isolation by segmental ostial ablation. Heart Rhythm 2004;1:197 –202.
6. Melby SJ, Lee AM, Zierer A, Kaiser SP, Livhits MJ, Boineau JP. Cardiac swelling
associated with linear radiofrequency ablation in the atrium. J Interv Card Electrophysiol 2001;5:159 – 66.
7. Wilber DJ, Pappone C, Neuzil P, De Paola A, Marchlinski F, Natale A et al. Comparison of antiarrhythmic drug therapy and radiofrequency catheter ablation in
patients with paroxysmal atrial fibrillation: a randomized controlled trial. JAMA
2010;303:333 –40.
8. Goldberger JJ. Atrial fibrillation ablation: closing the gaps. Heart Rhythm 2008;5:
1302 –3.
9. Melby SJ, Lee AM, Zierer A, Kaiser SP, Livhits MJ, Boineau JP et al. Atrial fibrillation
propagates through gaps in ablation lines: implications for ablative treatment of
atrial fibrillation. Heart Rhythm 2008;5:1296 – 301.
10. Chae S, Oral H, Good E, Dey S, Wimmer A, Crawford T et al. Atrial tachycardia
after circumferential pulmonary vein ablation of atrial fibrillation: mechanistic
insights, results of catheter ablation, and risk factors for recurrence. J Am Coll
Cardiol 2007;50:1781 –7.
11. Pappone C, Manguso F, Vicedomini G, Gugliotta F, Santinelli O, Ferro A et al. Prevention of iatrogenic atrial tachycardia after ablation of atrial fibrillation: a prospective randomized study comparing circumferential pulmonary vein ablation
with a modified approach. Circulation 2004;110:3036 –42.
12. Cappato R, Negroni S, Pecora D, Bentivegna S, Lupo PP, Carolei A et al. Prospective assessment of late conduction recurrence across radiofrequency lesions producing electrical disconnection at the pulmonary vein ostium in patients with
atrial fibrillation. Circulation 2003;108:1599 – 604.
13. Cheema A, Dong J, Dalal D, Marine JE, Henrikson CA, Spragg D et al. Incidence
and time course of early recovery of pulmonary vein conduction after catheter
ablation of atrial fibrillation. J Cardiovasc Electrophysiol 2007;18:387–91.
14. Goldberger JJ. Atrial fibrillation ablation: location, location, location. J Cardiovasc
Electrophysiol 2006;17:1271 –3.
15. Delacretaz E, Soejima K, Brunckhorst CB, Maisel WH, Friedman PL,
Stevenson WG. Assessment of radiofrequency ablation effect from unipolar
pacing threshold. Pacing Clin Electrophysiol 2003;26:1993 –6.
16. Eitel C, Hindricks G, Sommer P, Gaspar T, Kircher S, Wetzel U et al. Circumferential pulmonary vein isolation and linear left atrial ablation as a single-catheter
technique to achieve bidirectional conduction block: the pace-and-ablate approach. Heart Rhythm 2010;7:157 –64.
17. Sternick EB. Loss of pace capture on the ablation line: the quest for a more reliable endpoint for pulmonary vein isolation. Heart Rhythm 2010;7:331 –2.
18. Steven D, Reddy VY, Inada K, Roberts-Thomson KC, Seiler J, Stevenson WG et al.
Loss of pace capture on the ablation line: a new marker for complete radiofrequency lesions to achieve pulmonary vein isolation. Heart Rhythm 2010;7:323 –30.
19. Ranjan R, Kato R, Zviman MM, Dickfeld TM, Roguin A, Berger RD et al. Gaps in
the ablation line as a potential cause of recovery from electrical isolation and their
visualization using MRI. Circ Arrhythm Electrophysiol 2011;4:279 –86.
20. Datino T, Macle L, Chartier D, Comtois P, Khairy P, Guerra PG et al. Differential
effectiveness of pharmacological strategies to reveal dormant pulmonary vein
conduction: a clinical-experimental correlation. Heart Rhythm 2011;8:1426 – 33.