Preferential conduction patterns along the coronary sinus during atrial fibrillation in humans and their modification by pulmonary vein isolation.

Abstract

Correlation function analysis applied to endocardial electrograms has earlier been used for analysis of agreement between signals and direction of activation during atrial fibrillation (AF). This study was aimed at evaluating whether preferential activation patterns along the coronary sinus (CS) exist in patients with AF.
Twenty-seven patients (57 ± 10 years old) admitted for electrophysiological (EP) study (10 patients) and/or AF ablation (17 patients) were studied, 8 with permanent and 19 with persistent AF. Unipolar signals were recorded during 60 seconds from a 10-pole CS catheter during AF at baseline (BL) and after isolation of left and right pulmonary veins and after additional lines in the left atrium (LA) (End). Correlation function analysis was applied to signals from each pair of adjacent electrodes, and graphs of cumulated time delay were made to enable interpretation of direction of activation.
Correlation between paired signals was highest in the distal and middle parts of CS and lowest in the proximal CS. In 21 patients, correlation values greater than 0.8 between closely spaced electrodes suggested uniform propagation of the fibrillatory waves. In 18 of 21 patients, preferential conduction pattern along CS was seen. Of those, 15 patients had left-to-right conduction, and 3 had right-to-left conduction. During ablation, atrial fibrillation cycle length increased from 184 ± 32 milliseconds at BL to 193 ± 39 milliseconds after pulmonary vein isolation and 215 ± 39 milliseconds at the end of ablation (P = .03, BL vs End). Because of ablation, preferential conduction along CS changed in 4 patients from left to right at BL to simultaneous CS activation or right to left. In 1 of 3 patients with simultaneous activation at BL, the direction changed to right to left. No direction change was observed in any of the 3 patients with right-to-left activation at BL.
Atrial activation during AF exhibits a high degree of organization in distal and middle CS. Preferential conduction patterns observed in most patients may indicate either relatively dominant stable reentry circuits in the LA or activation spread from a focal source. The changes in preferential conduction during ablation of AF may reflect modification of AF substrate and indicate persistent right atrial sources not affected by ablation in the LA only.

Full text

Page 1

Preferential conduction patterns along the coronary sinus during atrial
fibrillation in humans and their modification by pulmonary vein isolation☆
Pyotr G. Platonov, MD, PhD, FESC,a,⁎Prashanthan Sanders, MBBS, PhD,b
Shashidhar, MD,bAnthony Brooks, PhD,bJari M. Tapanainen, MD, PhD,c
Fredrik Holmqvist, MD, PhD,aOle Kongstad, MD, PhD,aJonas Carlson, MSc, PhDa
aDepartment of Cardiology, Clinical Sciences, Lund University, and Center for Integrative Electrocardiology at Lund University (CIEL), Lund, Sweden
bCardiovascular Research Centre, Department of Cardiology, Royal Adelaide Hospital and the Discipline of Medicine, University of Adelaide,
Adelaide, Australia
cDivision of Cardiology, Helsinki University Hospital, Helsinki, Finland
Received 25 August 2010
AbstractIntroduction: Correlation function analysis applied to endocardial electrograms has earlier been
used for analysis of agreement between signals and direction of activation during atrial fibrillation
(AF). This study was aimed at evaluating whether preferential activation patterns along the coronary
sinus (CS) exist in patients with AF.
Methods: Twenty-seven patients (57 ± 10 years old) admitted for electrophysiological (EP) study (10
patients) and/or AF ablation (17 patients) were studied, 8 with permanent and 19 with persistent AF.
Unipolar signals were recorded during 60 seconds from a 10-pole CS catheter during AF at baseline
(BL) and after isolation of left and right pulmonary veins and after additional lines in the left atrium
(LA) (End). Correlation function analysis was applied to signals from each pair of adjacent electrodes,
and graphs of cumulated time delay were made to enable interpretation of direction of activation.
Results: Correlation between paired signals was highest in the distal and middle parts of CS and
lowest in the proximal CS. In 21 patients, correlation values greater than 0.8 between closely spaced
electrodes suggested uniform propagation of the fibrillatory waves. In 18 of 21 patients, preferential
conduction pattern along CS was seen. Of those, 15 patients had left-to-right conduction, and 3 had
right-to-left conduction. During ablation, atrial fibrillation cycle length increased from 184 ± 32
milliseconds at BL to 193 ± 39 milliseconds after pulmonary vein isolation and 215 ± 39
milliseconds at the end of ablation (P = .03, BL vs End). Because of ablation, preferential conduction
along CS changed in 4 patients from left to right at BL to simultaneous CS activation or right to left.
In 1 of 3 patients with simultaneous activation at BL, the direction changed to right to left. No
direction change was observed in any of the 3 patients with right-to-left activation at BL.
Conclusions: Atrial activation during AF exhibits a high degree of organization in distal and middle
CS. Preferential conduction patterns observed in most patients may indicate either relatively
dominant stable reentry circuits in the LA or activation spread from a focal source. The changes in
preferential conduction during ablation of AF may reflect modification of AF substrate and indicate
persistent right atrial sources not affected by ablation in the LA only.
© 2011 Elsevier Inc. All rights reserved.
Keywords:Atrial fibrillation; preferential conduction; correlation function analysis
Introduction
Development of atrial fibrillation (AF) requires an
initiating trigger and a substrate capable of maintaining the
arrhythmia. Catheter ablation, most commonly involving
electrical isolation of the pulmonary veins (PVs) by
producing lesion lines around the right and left PVs
(RPVs and LPVs, respectively), probably affects both the
trigger and the substrate.1,2However, attempts have been
Available online at www.sciencedirect.com
Journal of Electrocardiology 44 (2011) 157–163
www.jecgonline.com
☆Funding: The Swedish Heart-Lung Foundation, Nordic Research
Board (NordForsk), and Governmental funding of Clinical Research within
the Swedish NHS, Donation funds at Lund University Hospital.
⁎Corresponding author. Department of Cardiology, Lund University
Hospital, SE-221 85 Lund, Sweden.
E-mail address: Pyotr.Platonov@med.lu.se
0022-0736/$ – see front matter © 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.jelectrocard.2010.11.004

Page 2

made to limit the extent of the lesions by improving the
clinical efficacy of segmental PV ablation3or by targeting
focal sources that presumably drive the arrhythmia once it
is initiated.4,5These focal sources can be currently
identified as sites with high-frequency activity assessed
using spectral analysis5or continuous fractionated atrial
electrograms (CFAEs).4However, the exact mechanisms
underlying CFAE have not been fully elucidated, and its
ablation is commonly performed together with more
extensive circumferential PV isolation.6,7It is likely that
AF ablation procedure affects both the focal sources and
the way electrical activation travels across the atria.
Therefore, identification of preferential conduction might
indicate the origin of the activation and potentially be used
for guiding ablation.
Our group has developed a method aimed at the
identification of the preferential conduction pattern and
direction of propagation of fibrillatory waves during AF. It
is based on correlation function analysis between unipolar
endocardial electrograms.8In the pilot study, we were able
to document the preferential conduction along the
coronary sinus (CS) in 5 patients with AF, all of whom
presented left-to-right propagation of the fibrillatory
waves.8The current study was planned to further evaluate
the method and explore whether preferential direction of
propagation of fibrillatory waves in the CS can be
identified in a larger group of patients with different
types of AF and whether AF ablation procedure affects the
preferential direction of activation thus exploring its
potential to guide AF ablation procedure.
Materials and method
Patients
Twenty-seven consecutive patients admitted for clinically
motivated invasive electrophysiological (EP) study and/or
AF ablation comprised the study population (mean age, 57 ±
10 years; 23 male). Eight patients had permanent AF with
duration longer than 12 months, and 19 had persistent AF of
shorter duration. To be included in the study, the patients had
to have AF at BL. Most patients (13 of 27) were taking class
IC or class III antiarrhythmic drugs at the time of the
procedure. Three patients were treated with amiodarone, 7
with flecainide, and 7 with sotalol. The size of the left atrium
(LA) was assessed before the ablation using transthoracic 2-
dimensional echocardiography in a parasternal plane (mean
LA diameter, 43 ± 5 mm). The study complies with the
Declaration of Helsinki and appropriate permissions of the
local ethics committees.
Recordings and data acquisition
Ten unipolar endocardial electrograms and a standard 12-
lead electrocardiogram (ECG) were acquired simultaneously
using a BARD LabSystem Duo, C.R. BARD Inc., BARD
Electrophysiology Division, (Lund and Adelaide) or Prucka
CardioLab EP System, GE Medical Systems, (Helsinki). The
electrograms were acquired using a 10-pole catheter
(Dynamic XT 201-101; Bard Electrophysiology, Lowell,
MA) that was routinely placed in CS, with pole 10 near the
CS orifice and pole 1 (the tip of the catheter) in the distal CS.
The electrodes on the catheter were 1-mm wide and
organized in pairs, with a 5-mm space between the pairs
and 2 mm between the electrodes in a pair. Thus, a distance
of 40 mm was covered by the electrode. The signals were
sampled using identical settings with a 16-bit resolution and
a sampling frequency of 1 kHz per channel. All recordings
were at least 60 seconds in length. The 10 electrograms and
lead V1of the surface ECG were transferred to a GNU/Linux
workstation, where all subsequent analyses were made using
custom software written in Matlab R12 (The MathWorks,
Inc, Natick, MA).
Signal acquisition was performed at BL in all patients for
assessment of preferential conduction during clinical AF. In
17 patients with persistent AF, preferential conduction was
assessed during stepwise AF ablation procedure,6with
intraprocedural AF conversion as an end point. Ablation
procedure was limited to LA and included Lasso Biosense
Webster, catheter–guided PV isolation, linear ablation in
LA, and electrogram-based ablation aimed at sites showing
CFAE. Signals were acquired at BL, after LPV isolation,
after RPV isolation, and after additional lines and CFAE
ablation in the LA (End).
Data analysis
The technique applied for data analysis has been
described in detail earlier.8In short, the ventricular
components in the endocardial electrograms were identified
from surface ECG and removed from the electrograms by
linear interpolation. To assess whether unipolar signals
belonged to the same activation wave front and whether
analysis of preferential direction of its propagation was
feasible, the correlation function analysis was further
applied beat-by-beat to the signals from each of the 9
pairs of electrodes (CS1-2, CS2-3, CS3-4, etc), and the
maximum correlation (ρmax) as well as the corresponding
time delay between the signals from adjacent electrode
pairs was calculated (Fig. 1A-B). ρmaxWas used in this
study as a measure of similarity between unipolar
electrogram morphologies. Similar to the atrial rhythms
with organized and uniform propagation of depolarization
wave fronts along CS, that is, sinus rhythm and atrial
tachycardias,8the propagation of the fibrillatory waves
between CS electrodes was considered uniform if ρmaxN
0.6 (high degree of similarity).
To assess the direction of propagation of the fibrillatory
waves, cumulated time-delay graphs were made based on
the mean pairwise time delays between the adjacent
electrodes during the entire 60-second long recording.
With the applied definition of electrode-numbering, a
positive time shift indicated propagation from the pole
with the lower number to the one with the higher number
(ie, left to right or from distal CS toward its orifice),
whereas a negative time shift denoted propagation from
higher to lower numbered poles (ie, right to left or from CS
orifice toward its distal part). In the case of uniform
conduction from right to left, all values of cumulated time
158P.G. Platonov et al. / Journal of Electrocardiology 44 (2011) 157–163

Page 3

delay would be negative, whereas conduction from left to
right would result in positive values (Fig. 1C).
In all subjects, mean atrial fibrillatory frequency in the CS
was calculated.
Statistical analysis
Data are presented as means ± SD. In addition to the
calculation of the maximum correlation as described above,
the unpaired Mann-Whitney U test was used for comparison
of atrial fibrillatory frequency between subgroups of patients
with AF.
Results
Maximal correlation values between the unipolar signals
in CS
Correlation function analysis applied to each electrode
pair revealed that the agreement between paired signals
measured as maximal correlation value was the highest in the
distal and middle parts of CS (catheter poles 1-8). In this
region, 21 patients presented ρmax N 0.8 between the
electrodes separated by a 2-mm distance and ρmaxN 0.6
between the electrodes separated by a 5-mm space. In the
proximal CS, agreement between signals was poor, and ρmax
value generally was below 0.8 between the closely spaced
electrodes CS9-10.
Five patients had an “area” of low agreement between the
signals in the middle of the CS catheter (CS4-5, CS5-6, and
CS6-7), whereas showing ρmax N 0.6 in the remaining
proximal and distal electrode pairs. With these low
correlation values between the electrode pairs in the middle
CS, it was not reasonable to consider those subjects as
having uniform propagation along the CS, and these 5
patients were therefore excluded from analysis of the
preferential direction of activation. One more patient was
excluded because of extremely low ρmaxb 0.6 in 5 of 9
electrode pairs.
Preferential conduction patterns along the CS at BL
After exclusion of the 6 patients described above, the
preferential conduction pattern was assessed in the remain-
ing 21 patients. Of those, most (15) presented with left-to-
right pattern, that is, preferential conduction from the distal
to the proximal electrode pairs of the CS catheter. Three
patients had dominant right-to-left conduction, and 3
presented with nearly simultaneous activation of CS without
any preferential conduction (Fig. 2).
No differences between patients in relation to their sex,
clinical types of arrhythmia (permanent vs persistent), or use
of antiarrhythmic drugs were observed about the degree of
similarity between unipolar electrograms or the preferential
direction of propagation of fibrillatory wave fronts (data
not shown).
Effect of the AF ablation on the preferential
conduction pattern
During ablation procedure performed in 17 patients, AF
was terminated in 4 of them. All of these 4 patients had left-
to-right preferential conduction along CS at BL.
Of 17 patients admitted for ablation, 11 had left-to-right
conduction in the CS at BL. Of them, the shift in CS
conduction occurred after the encircling of the left superior
PV in 2 patients (case A, Fig. 3), after roofline ablation in 1,
and after mitral isthmus line in 1. Of 3 patients with
Fig. 1. Application of the correlation function analysis to the endocardial recordings. A, Ten unipolar signals recorded from a 10-polar catheter positioned in the
CS during AF. Gray areas correspond to the cancelled ventricular signals. B, Bar diagram illustrating maximal correlation (ρmax) between signals recorded from
adjacent electrodes. C, Cumulated time delay from the distal to the proximal electrode in the CS. Positive values of time delay (as in the shown example)
correspond to the preferential propagation of fibrillatory waves from CS1 toward CS10, that is, left to right.
159P.G. Platonov et al. / Journal of Electrocardiology 44 (2011) 157–163

Page 4

simultaneous activation in CS, 3 have shown conduction
change from right to left after isolation of the left common
PV (case B, Fig. 3).
In total, 5 of 17 patients presented with the shift in CS
conduction because of ablation procedure. In 3 of them, the
shift in conduction resulted in intraprocedural conversion of
AF to sinus rhythm. Only 1 of 12 patients who did not
demonstrate shift in CS conduction converted to sinus
rhythm during ablation procedure.
None of 3 patients with right-to-left conduction at BL has
presented with any change in conduction direction because
of ablation in the LA.
Atrial fibrillatory frequency in CS
Median CS atrial fibrillatory frequency in the total
population was 5.4 Hz, range, 3.6 to 6.9 Hz. In individual
patients, the variation of fibrillatory frequency recorded in 9
electrode pairs along the CS catheter was low, with median
coefficient of variance being 6.5%, range, 1% to 28%.
Patients with permanent AF had higher CS fibrillatory
frequency compared to the patients with persistent AF (6.1 ±
0.6 Hz vs 5.0 ± 0.6 Hz; P = .002). Because of ablation
performed in 17 patients, atrial fibrillatory frequency in the
CS decreased from 5.4 ± 0.9 Hz at BL to 5.1 ± 1.0 Hz after
LPV and RPV isolation and further to 4.6 ± 0.8 at the end of
the procedure (P = .03 BL vs End).
In the subgroup of patients where analysis of preferential
conduction was possible, no relation between CS atrial
fibrillatory frequency and the direction of preferential
conduction could be documented.
Discussion
One or more fibrillatory waves in the CS?
The method used in this study, correlation function
analysis, is a mathematically simple way of finding the
degree of similarity and time delay between 2 signal
sequences. The use of a correlation enables a simple
comparison between 2 analyses and also allows a result to
be “perfect” if the value of 1 is achieved. Unipolar
electrograms provide a richer signal morphology that is
associated with greater directional information9compared to
bipolar signals, and therefore, correlation analyses using
these signals should provide significant information
concerning wave front(s) consistency and AF complexity.
High correlation values together with similar atrial fibrilla-
tory frequency in different CS regions would support the
explanation that the signals recorded for the adjacent
123456789 10
0
10
20
30
40
50
Cumulated time delay [ms]
CS catheter pole
Left-to-right
123456789 10
-30
-20
-10
0
10
20
Cumulated time delay [ms]
CS catheter pole
12345678910
CS catheter pole
Right-to-left
DM
EE6567B
abl
-30
-20
-10
0
10
20
Cumulated time delay [ms]
No preferential conduction
hhm
MH6610B
tml
A
B
C
Fig. 2. Preferential conduction in the CS during AF illustrated by the graphs
of cumulated time delay from 21 patients who presented with a high degree
of similarity between the signals from adjacent electrodes recorded from the
catheter in the CS. A, Left-to-right conduction observed in most patients
(15). B, Right-to-left conduction. C, The lack of preferential conduction in
the CS.
Fig. 3. Changes in the preferential CS conduction during PV isolation. Case
A, Left-to-right conduction transforms to the simultaneous CS activation
after LPV isolation and accompanied by increase in AFCL. Case B,
Simultaneous CS activation transforms into the right-to-left propagation
after LPV isolation and accompanied by increase in AFCL. AFCL indicates
atrial fibrillation cycle length.
160P.G. Platonov et al. / Journal of Electrocardiology 44 (2011) 157–163

Page 5

electrodes belong to the same fibrillatory wave recorded
along 2 or more recording points in the CS. The lack of
agreement between the unipolar signals recorded in the
adjacent electrodes observed in 6 patients would mean that
the signals recorded in the CS do not belong to the same
wave front, and therefore, any further analysis of preferential
conduction is not feasible. When correlation function
analysis was applied earlier to cardiac rhythms characterized
by uniform conduction and known source and direction of
activation (eg, sinus rhythm and atrial tachycardias),
correlation value was, as a rule, higher than 0.6 and even
higher between the closely spaced electrodes (ρmaxN 0.8 in
CS1-2, CS3-4, etc).8It istherefore likely that,in applying the
same cutoff correlation values to the CS signals during AF,
the data in the present study can be explained by the
repetitive propagation of the more or less stable-over-time
fibrillatory wave front. In that regard, the observation of
generally higher correlation values in the distal CS compared
to that of the proximal CS is well in agreement with earlier
reported data showing that the lateral LA wall expresses
higher degree of organization compared to the LA wall
adjacent to the atrial septum.10
Direction of propagation of fibrillatory waves in the CS
Different degrees of organization of atrial activation
during AF have been reported earlier.11-13In the present
study, we used the catheter placed in the CS as a simple
model to test whether our preliminary findings on 5 patients
with AF could be reproduced in a larger material without the
need for additional catheterization prolonging the clinically
motivated ablation procedure.
The results illustrate that most patients with AF
demonstrate uniform preferential conduction pattern in the
CS, which is supported by the nearly identical values of atrial
fibrillatory frequency recorded from the different electrode
pairs of the CS catheter with low coefficient of variance
between the pairs being 10% or less in 21 patients. This may
indicate the presence of a relatively stable reentry circuit in
the LA or activation spread from a focal source. In the latter
case, it is not surprising that most study subjects had left-to-
right conduction during AF because the focal sources driving
AF are thought to be located mostly in the LA.5Recent data
provide evidence that, in patients with persistent AF, the
sources are, in fact, a reentrant and may also be located
outside PV ostia.14
Given the way the catheter was placed in the CS in our
study, the left-to-right conduction pattern may be explained
by the activation approaching the registration points from the
LPVs known as a common source of fibrillatory activity.
Likewise, the right-to-left conduction pattern observed in the
minority of study subjects may be explained by the
activation coming from interatrial septum or right atrium
via some of the earlier-described posterior or inferior
interatrial connections.15-17All of these structures are
known as potential sources of AF.18-20
As seen in Fig. 2A and B, recordings of right-to-left
conduction have a lower cumulated time delay than that of
left-to-right conduction. This may be explained by either the
impulse reaching the catheter at a greater angle or by a
greater conduction velocity. Although the angle may be the
most likely explanation in this case, the method cannot
distinguish between the two.
The lack of preferential conduction pattern is more
difficult to explain. We cannot exclude that this is simply a
reflection of the true lack of organization of fibrillatory
waves in the LA. The same cumulated time-delay graph
showing nearly simultaneous activation of registration points
in the CS could, however, be a result of the activation wave
fronts approaching CS from above, that is, from the RPVs
ostia or posterior LA wall. The high degree of similarity
between the unipolar signals recorded from adjacent
electrodes in these subjects supports this explanation.
One should keep in mind, of course, that activation
sequence in the CS may not truly reflect the direction of
fibrillatory waves in the LA in all patients. The possibility of
dissociation of electrical activation between the CS and the
LA during AF has earlier been documented in humans by
Ndrepepa et al,21who performed simultaneous mapping of
the LA and CS using multielectrode technique in a series of
42 patients with atrial tachyarrhythmias and reported the
presence of dissociation between the CS and LA activity in
only 9 patients with AF. The electrical dissociation is
presumably caused by the fact that CS is connected with the
LA musculature via discrete and variable muscular bridges
and not continuously on its entire length.15,22Being
heterogeneous, these connections can certainly contribute
to the electrical complexity independently from the global
AF complexity and has to be appreciated when interpreting
behavior of fibrillatory wave fronts recorded in the CS.
Preferential conduction in CS during AF ablation
In 5 of 17 patients who underwent AF ablation, we
observed a change of preferential conduction during the
ablation procedure. Every time when the direction of
preferential activation changed, the presumable source of
activation moved rightward, that is, left to right changed to
either simultaneous activation of electrode poles in the CS or
right-to-left propagation along CS. In cases when simulta-
neous activation was recorded at BL, it changed to right-to-
left wave front propagation along CS. In 2 of the 5 patients,
the conduction change occurred after ablation in the sensitive
areas that can potentially contain muscular bridges connect-
ing the CS with inferior LA wall, that is, encircling of the left
common PV and mitral isthmus in the vicinity of the distal
CS. Although ablation of these “entries” in the CS
musculature may serve as a sole explanation of the
conduction change in these patients, the other 3 patients
demonstrated conduction change after ablation of the remote
areas such as left superior PV isolation or left atrial roofline.
Although the number of these observations is small, it is
tempting to speculate that, in those subjects, ablation in the
LA affects dominating reentry circuits and shifts them away
from LPV and, possibly, to the right atrium when the
direction of preferential propagation becomes right to left.
Interestingly, none of the patients with right-to-left
propagation at BL has shown any change in preferential
161P.G. Platonov et al. / Journal of Electrocardiology 44 (2011) 157–163

Page 6

conduction pattern during ablation and has not converted to
sinus rhythm because of ablation either. The data processing
was performed off-line, and because no ablation in the right
atrium was performed after the completed procedure in the
LA, we have not been able to verify whether right-sided
procedure would further affect the preferential direction of
CS activation and/or result in AF conversion.
Clinical implications
Identification and ablation of driving rotors that maintain
AF are an attractive approach aimed at tailoring the ablation
procedure to the EP mechanisms playing a role in a given
patient. Recently, CFAEs have been proposed as an indicator
of EP substrate of AF,4,23and their ablation, in limited
studies, was shown to be a promising component of the
ablation procedure.24A specially designed catheter with
radiating spines for high-density mapping has also been
developed and applied for localization of focal sources
during ablation procedure.25
The method described in this manuscript is aimed at the
identification of the preferential conduction pattern during
AF along the axis of a diagnostic catheter and is an
alternative technique that can potentially be applied during
ablation procedures. If proven successful, it would improve
localization of focal sources driving AF and limit the extent
of ablation. However, future studies using analysis of atrial
electrograms during ablation in the LA are needed to further
validate this concept.
In this population, we have not been able to establish a
link between the degree of similarity of atrial fibrillatory
signals or the direction of propagation of atrial activation and
the clinical course of the arrhythmia. As expected, the atrial
fibrillatory rate was higher in patients with permanent AF.
Study limitations
Our study was designed as a retrospective analysis of data
obtained during clinically motivated AF ablation procedures
that, at the time of the study conduct, did not routinely
involve right atrial ablation. The data on CS activation
change appeared because of an off-line analysis and thus
could not be used in guiding the ablation procedure in cases
when, for example, the right-to-left activation pattern
persisted at the end of ablation in the LA. Secondly, for
the purpose of the present study, CS musculature was
assumed to be a passive electrical conduit, although we
appreciate that, in some cases, AF drivers may exist within
the CS. The investigation of CS as a potential driver of AF
would, however, require extensive biatrial multipoint
frequency mapping to define sites of highest frequency
activation and was not a part of the routine AF ablation at the
participating sites.
Conclusion
Correlation function analysis applied to the unipolar
electrograms recorded from the CS during AF suggests that
uniform propagation of atrial activation along the CS is a
common finding in patients with AF. Preferential conduction
patterns observed in most patients may indicate either
relatively dominant and stable reentry circuits or activation
spread from a focal source. The changes in preferential
conduction during ablation procedures observed in one third
of the patients may reflect modification of AF substrate in the
LA, whereas the lack of change in patients with right-to-left
conduction at BL may indicate persistent right atrial sources
not affected by ablation in the LA only. Further evaluation of
the method during ablation procedures is warranted to
investigate its value in the localization of arrhythmogenic
sources during AF.
Acknowledgments
The study was supported by research grants from The
Swedish Heart-Lung Foundation, Torsten Westerström
Foundation, Nordic Research Board (NordForsk), and
governmental funding of clinical research within the
Swedish National Healthcare System (NHS). Dr Sanders
and Dr Brooks are supported by the National Heart
Foundation of Australia.
References
1. Pappone C, Rosanio S, Augello G, et al. Mortality, morbidity, and
quality of life after circumferential pulmonary vein ablation for atrial
fibrillation: outcomes from a controlled nonrandomized long-term
study. J Am Coll Cardiol 2003;42:185.
2. Marrouche NF, Martin DO, Wazni O, et al. Phased-array intracardiac
echocardiography monitoring during pulmonary vein isolation in
patients with atrial fibrillation: impact on outcome and complications.
Circulation 2003;107:2710.
3. Yamada T, Murakami Y, Okada T, et al. Can segmental pulmonary
vein ablation reduce the recurrence of atrial fibrillation when using a
higher RF power, larger tip electrode catheter, and additional RF
deliveries?: the limitations of point-by-point RF ablation. Int Heart J
2006;47:219.
4. Nademanee K, McKenzie J, Kosar E, et al. A new approach for catheter
ablation of atrial fibrillation: mapping of the electrophysiologic
substrate. J Am Coll Cardiol 2004;43:2044.
5. Sanders P, Berenfeld O, Hocini M, et al. Spectral analysis identifies
sites of high-frequency activity maintaining atrial fibrillation in humans.
Circulation 2005;112:789.
6. O'Neill MD, Jais P, Takahashi Y, et al. The stepwise ablation approach
for chronic atrial fibrillation—evidence for a cumulative effect. J Interv
Card Electrophysiol 2006;16:153.
7. Calkins H, Brugada J, Packer DL, et al. HRS/EHRA/ECAS expert
consensus statement on catheter and surgical ablation of atrial
fibrillation: recommendations for personnel, policy, procedures and
follow-up. A report of the Heart Rhythm Society (HRS) Task Force on
catheter and surgical ablation of atrial fibrillation. Heart Rhythm
2007;4:816.
8. Carlson J, Santos S, Platonov PG, Rasmussen OK, Johansson R, Olsson
SB. Left atrial conduction along the coronary sinus during ectopic atrial
tachycardia and atrial fibrillation: a study using correlation function
analysis. J Cardiovasc Electrophysiol 2003;14:S148.
9. Simmers TA, Hauer RN, Wever EF, Wittkampf FH, Robles de Medina
EO. Unipolar electrogram models for prediction of outcome in
radiofrequency ablation of accessory pathways. Pacing Clin Electro-
physiol 1994;17:186.
10. Ndrepepa G, Karch MR, Schneider MA, et al. Characterization of
paroxysmal and persistent atrial fibrillation in the human left atrium
during initiation and sustained episodes. J Cardiovasc Electrophysiol
2002;13:525.
11. Karch MR, Ndrepepa G, Zrenner B, et al. Simultaneous multisite
endocardial mapping of sustained and non-sustained atrial fibrillation in
humans. J Invasive Cardiol 2003;15:257.
162P.G. Platonov et al. / Journal of Electrocardiology 44 (2011) 157–163

Page 7

12. Gaita F, Calo L, Riccardi R, et al. Different patterns of atrial activation
in idiopathic atrial fibrillation: simultaneous multisite atrial mapping in
patients with paroxysmal and chronic atrial fibrillation. J Am Coll
Cardiol 2001;37:534.
13. Roithinger FX, SippensGroenewegen A, Karch MR, Steiner PR, Ellis
WS, Lesh MD. Organized activation during atrial fibrillation in man:
endocardial and electrocardiographic manifestations. J Cardiovasc
Electrophysiol 1998;9:451.
14. Atienza F, Almendral J, Moreno J, et al. Activation of inward rectifier
potassium channels accelerates atrial fibrillation in humans: evidence
for a reentrant mechanism. Circulation 2006;114:2434.
15. Mitrofanova L, Ivanov V, Platonov PG. Anatomy of the inferior
interatrial route in humans. Europace 2005;7(Suppl 2):49.
16. Ho SY, Sanchez-Quintana D, Cabrera JA, Anderson RH. Anatomy of
the left atrium: implications for radiofrequency ablation of atrial
fibrillation. J Cardiovasc Electrophysiol 1999;10:1525.
17. Dong J, Zrenner B, Schreieck J, et al. Catheter ablation of left atrial
focal tachycardia guided by electroanatomic mapping and new
insights into interatrial electrical conduction. Heart Rhythm 2005;
2:578.
18. Ino T, Miyamoto S, Ohno T, Tadera T. Exit block of focal repetitive
activity in the superior vena cava masquerading as a high right atrial
tachycardia. J Cardiovasc Electrophysiol 2000;11:480.
19. Tsai CF, Tai CT, Hsieh MH, et al. Initiation of atrial fibrillation by
ectopic beats originating from the superior vena cava: electrophysio-
logical characteristics and results of radiofrequency ablation. Circula-
tion 2000;102:67.
20. Yamada T, Murakami Y, Okada T, Murohara T. Focal atrial fibrillation
associated with multiple breakout sites at the crista terminalis. Pacing
Clin Electrophysiol 2006;29:207.
21. Ndrepepa G, Zrenner B, Schneider MA, et al. Dissociation between
coronary sinus and left atrial conduction in patients with atrial
fibrillation and flutter. J Cardiovasc Electrophysiol 2001;12:623.
22. Chauvin M, Shah D, Haissaguerre M, Marcellin L, Brechenmacher C.
The anatomic basis of connections between the coronary sinus
musculature and the left atrium in humans. Circulation 2000;101:647.
23. Rostock T, Rotter M, Sanders P, et al. High-density activation mapping
of fractionated electrograms in the atria of patients with paroxysmal
atrial fibrillation. Heart Rhythm 2006;3:27.
24. Schmitt C, Estner H, Hecher B, et al. Radiofrequency ablation of
complex fractionated atrial electrograms (CFAE): preferential sites of
acute termination and regularization in paroxysmal and persistent atrial
fibrillation. J Cardiovasc Electrophysiol 2007;18:1039.
25. Sanders P, Hocini M, Jais P, et al. Characterization of focal atrial
tachycardia using high-density mapping. J Am Coll Cardiol
2005;46:2088.
163P.G. Platonov et al. / Journal of Electrocardiology 44 (2011) 157–163