Download Non-invasive detection of conduction pathways

CLINICAL RESEARCH
Europace (2009) 11, 169–177
doi:10.1093/europace/eun335
Electrophysiology and Ablation
Non-invasive detection of conduction
pathways to left atrium using
magnetocardiography: validation by
intra-cardiac electroanatomic mapping
Raija Jurkko1,2*, Ville Mäntynen 2,3, Jari M. Tapanainen 1,2, Juha Montonen 2,
Heikki Väänänen 3, Hannu Parikka 1,2, and Lauri Toivonen 1,2
1
Department of Cardiology Helsinki University Central Hospital, FI-00290 Helsinki, Finland; 2BioMag Laboratory HUSLAB, Helsinki University Central Hospital, Helsinki,
Finland; and 3Laboratory of Biomedical Engineering, Helsinki University of Technology, Espoo, Finland
Received 13 September 2008; accepted after revision 11 November 2008; online publish-ahead-of-print 13 December 2008
Aims
Alteration in conduction from right to left atrium (LA) is linked to susceptibility to atrial fibrillation (AF). We examined whether different inter-atrial conduction pathways can be identified non-invasively by magnetocardiographic
mapping (MCG).
.....................................................................................................................................................................................
Methods
In 27 patients undergoing catheter ablation of paroxysmal AF, LA activation sequence was determined during sinus
and results
rhythm using invasive electroanatomic mapping. Before this, 99-channel magnetocardiography was recorded over
anterior chest. The orientation of the magnetic fields during the early (40 –70 ms from P onset) and later part
(last 50%) of LA depolarization was determined using pseudocurrent conversion. Breakthrough of electrical activation to LA occurred through Bachmann bundle (BB) in 14, margin of fossa ovalis (FO) in 3, coronary sinus
ostial region (CS) in 2, and their combinations in 10 cases by invasive reference in total of 29 different P-waves.
Based on the combination of pseudocurrent angles over early and late parts of LA activation, the MCG maps
were divided to three types. These types correctly identified the LA breakthrough sites to BB, CS, FO, or their combinations in 27 of 29 (93%) cases.
.....................................................................................................................................................................................
Conclusion
Magnetocardiographic mapping seems capable of distinguishing inter-atrial conduction pathways. Recognizing the
inter-atrial conduction pattern may assist in understanding the pathogenesis of AF and identifying the subgroups
for patient-tailored therapy.
----------------------------------------------------------------------------------------------------------------------------------------------------------Keywords
Atrial fibrillation † Inter-atrial conduction † Electroanatomic mapping † Magnetocardiography
Introduction
Conduction disturbances in the atria have a role in the genesis and
maintenance of atrial fibrillation (AF).1,2 In the intact human heart,
the conduction from the right atrium (RA) to the left atrium (LA) is
known to occur through Bachmann bundle (BB), the rim of fossa
ovalis (FO), and the coronary sinus ostial region (CS).3 – 9 The
RA activation is oriented mainly from the right of the subject to
leftward down.4,6,10,11 Similar descending activation pattern has
been demonstrated in the LA. However, signal propagation in LA
can be also ascending and has then been related to block in
BB.11,12 The signal breakthrough sites in LA are in accordance
with anatomic muscle bundles;13 – 16 the number, location, and
thickness of which are largely variable. The anatomical bundles
are known to be variable also in healthy subjects,14,16 but the
knowledge of inter-atrial conduction is derived mainly from
patients with clinical arrhythmias. In order to investigate large
populations, more applicable non-invasive methods are needed.
Some electrocardiographic features, such as P-wave prolongation and morphological changes in the end of the P-wave,
* Corresponding author. Tel: þ358 9 4717 2442, Fax: þ358 9 4717 4574, Email: [email protected]
Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2008. For permissions please email: [email protected].
170
have been proposed to represent inter-atrial conduction block and
vulnerability to AF.12,17,18 Recently, three vectorcardiographic atrial
wave patterns were related to distinct inter-atrial conduction pathways.9 Magnetocardiography (MCG) is a non-invasive method
complementary to ECG to examine cardiac electromagnetic
activity. Essential to electrical and magnetic fields is a 908 spatial
angle between the fields.19 In MCG, the currents tangential to Bz
component, which is perpendicular to the sensor surface and
anterior chest, yield the strongest signal whereas ECG is sensitive
to radial currents. Magnetocardiography may therefore show deviation from normal direction of depolarization and repolarization
differently compared with ECG.19 Consequently, currents tangential to the chest surface, as most currents in the atrial walls are,
might be better detectable by MCG than by ECG. Magnetocardiography is also less affected by conductivity variations caused by the
lungs, muscles, and skin.19
Magnetocardiography has been accurate in detecting ventricular
arrhythmia substrate and localizing cardiac electrical sources,20,21
and can also be used to analyse the direction of cardiac activation
sequences.22 Preliminary results of MCG mapping of atrial signal
propagation during sinus rhythm suggested more variation in
atrial activation pattern between AF patients than healthy subjects.23 Differences were seen mainly in the later parts of the
atrial complex and were hypothesized to reflect variation in the
propagation of LA activation due to differences in inter-atrial conduction. To test this hypothesis, we examined the relationship
between atrial activation patterns derived from MCG mapping
and invasive electroanatomic activation maps.
Methods
Study population
The study included 27 patients undergoing electrophysiological study
prior to catheter ablation therapy of paroxysmal AF. The presence
of structural heart disease was assessed by clinical, ECG, and cardiac
ultrasound examinations. Anti-arrhythmic medication was discontinued at least for five half-life times except in five patients.
The study was approved by the Ethical Review Board of the institute, and written informed consent was obtained from all patients.
Intra-cardiac electroanatomic mapping
Electroanatomic mapping was performed using an electroanatomical
mapping system (CARTOwXP system, Biosense Webster, Inc.,
Diamond Bar, CA, USA) with either a 7 Fr Navi-Star or ThermoCool
catheter (Biosense Webster). A decapolar 6 Fr diagnostic catheter was
placed in the CS. Three-dimensional isochronal activation maps were
generated in sinus rhythm, gated to a stable CS reference signal.24,25
The number of registered points was 112 + 37 in LA and 71 + 18
in RA. All points and respective ECG and intra-cardiac recordings in
the maps were visually inspected. Local activation time was determined
as the maximum or minimum of the first sharp deflection with an
absolute value over 0.1 mV of the bipolar signal at the distal electrode
pair of the catheter.9,26
P-wave morphology was examined to ensure sinus origin. The point
was rejected if the beat was ectopic, the intra-cardiac signal was
,0.1 mV in amplitude, or signs of catheter instability were
seen.5,6,9,10,24 – 26 If more than one sinus P-wave morphology were
seen in a recording, each was analysed separately. When double
R. Jurkko et al.
potentials were found, the first was used except when it was a low
frequency far-field potential.9 All intra-cardiac electrogram data were
gathered before starting the ablation procedure.
Electroanatomic maps (EAMs) of atrial activation were reconstructed by applying interpolated colour code adjustments of local activation times on recorded anatomic shape. The area of first activation in
the LA was determined. Conduction through BB was assumed when
the earliest activation was in upper third of LA, superior and leftward
from the upper right pulmonary vein. Conduction through the rim of
FO was assumed when earliest activation was within the middle third
of LA, around the trans-septal puncture site. Conduction through CS
region was assumed when the earliest activation was in the lowest
1 cm of LA, with activation front directing cranially. More than one
electrical breakthrough site was regarded to exist when distinct conduction sites were activated within 15 ms and were separated by
areas showing later activation.10 According to the observed breakthrough sites, the patients were separated into BB, CS, FO, and combined (multisite activation) groups. The LA activation sequence was
examined using both activation and propagation maps, and the direction of signal propagation was assessed visually. All electroanatomic
signals were examined by two readers, and in case of discordance in
breakthrough sites, consensus was reached after consulting an electrophysiologist experienced in cardiac mapping.
The onset of atrial activation was determined from the 12-lead ECG
as the earliest time point where the signal could be separated from the
baseline. The time from onset to the latest intra-cardiac LA activation
was defined as the total atrial activation time. Total LA and RA activation times were determined from intra-cardiac recordings.
Magnetocardiographic method
Magnetocardiography recordings were performed in a magnetically
shielded room (ETS-Lindgren Euroshield Oy, Eura, Finland) using a
multichannel cardiomagnetometer (Elekta Neuromag Ltd, Helsinki,
Finland) equipped with 33 triple sensor dc-SQUID units on a slightly
curved surface with a diameter of 30 cm. In each unit, a magnetometer
is overlaying two orthogonal planar gradiometers, the magnetometer
coil direction being perpendicular to the sensor array (z-axis). The
system measures the magnetic field Bz component and the spatial
change of this component, the planar gradients, d(Bz)/dx and d(Bz)/dy.
Magnetocardiography was recorded over anterior chest, the centre
of the sensor array positioned 15 cm below the jugular notch and 5 cm
left from the midsternal line, in sinus rhythm over 7 min (Figure 1). Simultaneously, the limb leads of the standard ECG were recorded. Analogue signal pass-band was 0.03 – 300 Hz and sampling frequency of
analogue-to-digital conversion was 1000 Hz. The data were averaged
using atrial wave template and maximum cross correlation.27 Atrial
ectopic beats and excessively noisy sinus beats were rejected. If two
distinct sinus P-wave morphologies were present, both were separately averaged. The onset and end of the P-wave as well as P-wave duration were automatically determined using 40 Hz high-pass filtering
and a computerized algorithm as described earlier.27
Time interval over the first 30 ms of the atrial complex was taken to
represent the early part of RA activation. The time interval over
40 – 70 ms after the beginning of atrial complex was chosen to
represent early part of LA activation and the latter half of the atrial
wave to represent the later part of LA activation. The selection of
time intervals was based on previous knowledge on the atrial activation
sequence.4 – 7,10,11
Integrals of the magnetic field Bz component over the defined time
intervals were interpolated at the sensor array plane using magnetic
multipole expansion.28 To characterize the orientation of the magnetic
field, a pseudocurrent conversion was used.29 The method is based on
171
MCG mapping of conduction to LA
Figure 1 Recording and analysis of atrial magnetic fields. (A) The sensor arrangement of a 33 U triple sensor (99-channel) magnetometer.
Superimposed on sensor array is the antero-posterior few of the left atrium by electroanatomic mapping. (B) Signal-averaged magnetic field
density on each magnetometer channel over cardiac cycle. The onset and end of atrial signal are determined automatically using filtering technique. (C) Spatial distribution of the magnetic field Bz component over the middle part of atrial complex interpolated from the measurement
using multipole expansion. The blue colour indicates flux out of the chest (2) and red colour flux into the chest (+). The step between two
consecutive lines is 200 ft. (D) Pseudocurrent map derived by rotating magnetic field gradients by 908. The red– yellow colour indicates the area
of the top 30% of strongest currents, and the large arrow indicates their mean direction. Zero angle direction is pointing from subject’s right to
left and positive clockwise.
rotating the estimated planar gradients of the Bz component by 908:
a~ ¼
@Bz
@Bz
e^ x e^ y ;
@y
@x
where ex and ey are the perpendicular unit vectors on the sensor
array plane. The resulting arrow map provides a zero-order approximation (pseudocurrent map) for the underlying electric currents. To
get a quantitative variable, the mean of the angles of the top 30% of
the strongest pseudocurrents was calculated, zero angle direction
pointing from subject’s right to left and positive clockwise. The
method is illustrated in Figure 1 and described in more detail in the
Appendix.
The MCG map orientations during 40– 70 ms after the onset of the
atrial complex and over its last 50% were compared with LA breakthrough areas in EAM in each individual. The pseudocurrent angles
in MCG maps were calculated for the whole patient group and for subgroups with different breakthrough sites. In addition, in six patients, the
MCG map orientation of the integral over first 30 ms of atrial depolarization complex was compared with EAM of the RA. The P-wave morphology in limb leads of the standard ECG recorded during both
mappings was used to capture similar atrial activity in each EAM and
MCG map pairs.22 An example of the EAM data and MCG data in
the same timescale is shown in Figure 2.
Statistical analysis
Continuous data are expressed as mean + SD and categorical variables as numbers and proportion of positive cases in groups. Differences between the groups were examined using Student’s t-test for
continuous and the x 2 test for discrete variables. Coefficient of variation was used to compare measurements obtained by two different
methods.
Angular data are expressed as mean angle and circular standard
deviation (CSD). Angular – angular correlation coefficient was used
to study the relationship of the MCG map orientations, and
Watson’s U 2 test was used for comparison between groups.30 A twotailed P-value of ,0.05 was considered statistically significant.
Results
Demographic and clinical data of the patients are presented in
Table 1. The mean age was 45 years and most patients were
male. All had paroxysmal AF and most (89%) had no structural
heart disease. In 2 patients, 2 different sinus P-wave morphologies
were included resulting in total of 29 LA maps. In six patients also
RA was mapped intra-cardially.
Electroanatomic maps
A single inter-atrial breakthrough site was observed in 19 of 29
maps (65%). The pathways were BB in 14 (48%), FO in 3 (10%),
and CS in 2 (7%) cases by the invasive reference. In the remaining
10 of 29 cases (34%), the activation occurred through more than
one pathway. Here, BB was included in six, FO in nine and CS in
seven cases.
The directions of activation fronts by EAM during early LA activation, evaluated visually, were mainly leftward down, i.e. descending, in the BB group, leftward up, i.e. ascending, in the CS group,
and more horizontal and variable in the FO and combined
groups. The main direction of activation front during RA activation
was mostly leftward down as illustrated in Figure 2.
The duration of total atrial activation by EAM was 117 + 12 ms
(Table 2). The earliest LA activation was detected 34 + 9 ms after
the onset of atrial complex. The duration of LA activation was
84 + 14 ms. In six cases available, the duration of RA activation
was 81 + 8 ms. Both atria were activated simultaneously for
49 + 12% of the total atrial activation time.
Magnetocardiographic maps
The duration of filtered atrial wave by MCG was 115 + 15 ms. The
durations are shown for groups formed according to different LA
breakthroughs in Table 2. The filtered atrial wave duration
and total activation time in EAM showed 5.1% coefficient of
variation.
172
R. Jurkko et al.
Figure 2 Example of atrial electroanatomic and magnetocardiographic mappings. Upper panel: electroanatomic map: isochronal activation
maps of the right atrium and left atrium in antero-posterior (left) and postero-anterior (right) projections shown with colour-coded timescale.
The step between two isochronal lines is 5 ms. The total atrial activation time in this case is 120 ms. The earliest right atrium activation is seen at
upper lateral-septal part of the atrium (red star). The earliest left atrium activation is in the middle inter-atrial septum and 5 ms later at the area
of Bachmann bundle and 10 ms later at the coronary sinus ostial area. The multisite activation pattern is best visible from the back—marked by
thin arrows, the upper representing Bachmann bundle and lower coronary sinus routes. The activation fronts are pointed using arrows and
symbols, dotted lines for right atrium, solid lines for initial left atrium, and dashed lines for later left atrium activation. Lower panel: magnetocardiographic map: pseudocurrent density maps representing the right atrium (first 30 ms), initial left atrium (40 – 70 ms), and later left
atrium (last 50% of the whole atrium) activation time. Maps represent slightly tilted frontal plane projections with horizontal line from subject’s
right to left. The red – yellow areas correspond to the top 30% of the pseudocurrent amplitudes, and their mean angle is indicated with yellow
arrows.
The pseudocurrent direction in MCG maps over the first 30 ms
of atrial complex, representing early RA activation, was mostly leftward down, with a mean angle of 438 (CSD 288). Over the time
interval of 40–70 ms from the onset of the atrial complex, the
mean angle was 398 (CSD 308). Over the time interval of last
50% of atrial complex, the mean angle was 38 (CSD 518). An
example of MCG maps over the selected time intervals of atrial
depolarization is shown in Figure 2.
When both the early and late LA MCG maps were viewed
together, three types of combinations emerged: Type 1 with
both maps showing pseudocurrent orientation leftward down,
Type 2 with the map over the 40 –70 ms orienting leftward
down and the map over last 50% of atrial signal orienting leftward
up, and Type 3 with both maps orienting leftward up. Examples of
these three types are illustrated in Figure 3.
Relationship between pseudocurrent
directions and left atrium
breakthrough sites
The activation fronts in MCG maps differed between subgroups allocated with regard to LA breakthrough sites in EAMs. Over the time
interval 40–70 ms during the atrial complex, the pseudocurrent
mean angle in the BB group pointed leftward down, more horizontally in the FO and combined groups, and leftward up in the CS
group, as indicated in Table 2 and illustrated in Figure 3. The
173
MCG mapping of conduction to LA
distribution of magnetic field orientation in the BB group (n = 14) was
significantly different from that in other groups (n = 15) (P , 0.02).
Overlapping of the multisite breakthrough group with other groups
decreased the separating power of this measure, as seen in Figure 4.
The direction of pseudocurrent angle over the latter half of atrial
complex was positive in all cases in BB group and in 2 cases of
combined pathways (BB & CS and BB & FO), but negative in all
other 13 maps. Also during this time interval, the distribution of
magnetic field orientation in the BB group was significantly different from that in other groups (P , 0.001).
Magnetic field orientations over the initial and later part of LA
activation showed high mutual correlation (angular– angular r =
0.89, P , 0.001) in cases with single breakthrough via BB or CS,
but differed in cases with FO and combined breakthroughs. The
combination of two LA MCG maps yielded three different types.
All 14 cases of solitary breakthrough via BB had Type 1 magnetic
field maps (Table 3). The three cases with solitary FO breakthrough had Type 2 maps and both cases with solitary CS breakthrough had Type 3 maps. In combined pathways (n = 10), Type
2 maps were found in eight cases and Type 1 maps in two cases.
Overall, by using the MCG map type as a criterion, the LA
breakthrough site was correctly identified to BB, CS, and FO or
combined pathways in 27 of 29 cases (93%). Only the solitary
breakthrough via FO could not be separated from the combined
pathways.
Discussion
Main findings
Table 1 Demographic and clinical features
Number of patients
Male/female
27
21/6
Age (years)
45 + 10 (range 20– 57)
Height (cm)
Weight (kg)
179 + 11
86 + 18
BMI
27 + 4
Age at diagnosis of AF (years)
Concomitant diseases
39 + 10
4
Arterial hypertension
1
Cardiomyopathy
Aortic valve regurgitation
1
1
Long QT syndrome
1
Cardiac ultrasonography
Left atrial diameter (mm)
Left ventricular end-diastolic
diameter (mm)
Left ventricular ejection fraction (%)
Class I or III anti-arrhythmic agent
at measurement
Invasive assessment of connections
to the left atrium
39 + 4
In concordance with earlier studies,3 – 6,9 most of our patients
showed conduction through BB, either as a solitary route or in combination with other routes. However, a remarkable 27% minority of
patients showed none of the LA activated through the BB route. This
is supported by an earlier non-contact mapping study where the BB
route was not observed in 12 of 21 patients7 as well as by a postmortem anatomic study where BB was not seen in half of the AF
patients or controls.16 The activation of LA through multiple conduction pathways was found in approximately one-third of our
patients, which is in the range of 10–43% as reported earlier.7,9 In
two of our patients, two activation routes alternated during
53 + 6
63 + 6
5/27 (19%)
Amiodarone
Flecainide
1
2
Disopyramide
2
The present study demonstrates that the variation in impulse
propagation to the LA during sinus rhythm can be assessed noninvasively by magnetocardiography. Using the pseudocurrent
pattern of the initial and later parts of LA signal, three distinct
LA breakthrough sites corresponding to anatomical areas of BB,
inter-atrial septum at the vicinity of FO, and the CS ostium
could be separated. Judged by pseudocurrent directions in MCG
maps, the signal propagation during the initial and later parts of
LA activation was directed leftward down in BB breakthrough,
upwards in CS breakthrough, and variably horizontal left when
breakthrough area was FO or multiple breakthroughs appeared.
Overall, the breakthrough sites could be classified correctly into
these categories by MCG in 27 of the 29 cases (93%).
The figures represent the number of study subjects or mean + SD. BMI, body
mass index; weight (kg)/height (m)2.
Table 2 Atrial activation times and magnetocardiographic pseudocurrent orientation in whole study population and in
subgroups formed according to left atrial breakthrough areas in electroanatomic mapping
All n = 29
BB n = 14
CS n = 2
FO n = 3
Combined n = 10
...............................................................................................................................................................................
Duration of total atrial activation in electroanatomic map (ms)
117 + 12
116 + 10
102 + 3
123 + 11
119 + 12
P-wave duration in MCG (ms)
115 + 15
111 + 13
101 + 6
120 + 18
121 + 18
LA map (40–70 ms) orientation (8)
LA map (last 50%) orientation (8)
+39 (30)
23 (51)
+57 (23)*
+38 (24)**
223 (11)
251 (1)
+21 (11)
240 (60)
+30 (20)
232 (41)
The figures represent mean + SD or mean and circular standard deviation (in parentheses). BB, Bachmann bundle; FO, margin of fossa ovalis; Combined, two or more
breakthrough areas; CS, coronary sinus ostium; LA, left atrium.
Statistical significances: *P , 0.02 and **P , 0.001 comparing BB group with other groups based on left atrial MCG map type.
174
R. Jurkko et al.
Figure 3 Electroanatomic and magnetocardiographic maps. (A) Isochronal activation electroanatomic maps during sinus rhythm in anteroposterior projection from four cases with different inter-atrial propagation routes. Left atrium is activated through Bachmann bundle, rim of
the fossa ovalis, coronary sinus ostium, and combination of Bachmann bundle, fossa ovalis, and coronary sinus routes. Red colour identifies
the earliest and purple colour the latest activation. The step between two isochronal lines is 5 ms. Activation breakthroughs are marked
with black circles. Pseudocurrent density maps representing (B) initial left atrial (40 – 70 ms) and (C) later left atrial (last 50%) activation
from respective cases. Maps represent slightly tilted frontal plane projections with horizontal line from subject’s right to left. The red –
yellow areas correspond to the top 30% of the pseudocurrent amplitudes and their mean angle, the magnetocardiographic map orientation
is indicated with large yellow arrows. The Types 1, 2, and 3 (at the bottom) refer to the classification based on pseudocurrent direction in
the two magnetic field maps.
recording indicating that temporary factors can modify the conduction pattern. A shift in an endocardial LA breakthrough site was also
demonstrated by Markides et al.7
The durations of whole atrial activation and LA activation
measured in the present study were comparable with the durations of 120 + 24 and 80 + 11 ms reported by Lemery et al.5 in
AF patients treated with catheter ablation, and slightly longer
than the duration of LA activation of 65 + 4 ms reported by
Markides et al.7 in lone AF patients. The LA activation started at
34 ms on average in our study, similar to earlier observations.5 – 7,11
Since RA activation is rather stable according to previous4 – 6,10,11,24 and present observations, variation during the
40 –70 ms from atrial onset apparently has the capability to
reveal different patterns of the superimposed LA activation. This
justifies the use of the middle part of atrial depolarization wave
to analyse different LA activation patterns.
Interpreting atrial activation by
magnetocardiographic mapping
In some recent works, interpolation of the current density maps,31
or independent component of multichannel magnetic field signal,32
on three-dimensional heart model have been used. However, the
two-dimensional presentation of the pseudocurrent angle utilized
in this study29,33 seems to represent adequately the main direction
of electrical signal propagation. The findings support the concept
that two-dimensional pseudocurrent map can provide an estimate
of the summation of real three-dimensional atrial currents and
their temporal propagation.29,33
It has been previously suggested that the LA breakthrough site is
reflected in the LA activation pattern.5,7,8,11 In the present study,
the LA activation fronts over the initial part of LA activation differed between BB, CS, and FO conduction pathway subgroups.
However, the FO activation route, as solitary or in combination,
could not be clearly distinguished based on the initial LA activation
alone. When the information over the latter half of atrial complex
was combined with that of initial LA, the MCG maps could be
divided into three different types, each of which was suggestive
to a certain breakthrough site. The types were specific to activation through BB and CS, but the activation through the margin
of FO seems to create less distinct activation pattern. Since FO
was involved in 8 of the 10 cases in the combined group, the activation through FO might confound assessment of conduction
through BB. The findings are comparable to orthogonal ECG by
which the single route activation through FO or BB could not be
separated from multisite activation including these pathways.9
175
MCG mapping of conduction to LA
Overall, the findings imply that non-invasive MCG mapping can
be utilized to assess conduction to the LA during sinus rhythm.
Although MCG has better localizing ability compared with electrocardiography,21 the observations nevertheless encourage to
develop computational analysis of data acquired by standard or
modified electrocardiographic lead sets.9,34,35 With more sophisticated applications, MCG technology might provide a new means of
non-invasive mapping of atrial arrhythmia foci and re-entrant
circuits.
Clinical associations
Figure 4 The pseudocurrent mean angles of 40 – 70 ms (x-axis)
and last 50% (y-axis) time intervals in the x – y plot. Each case is
labelled according to the left atrium breakthrough site in invasive
mapping: solid circle = Bachmann bundle, solid triangle = coronary sinus, open square = fossa ovalis, multiplication symbol =
Combination. In cases with a solitary conduction pathway, both
angles shift to less negative when moving from Bachmann
bundle to fossa ovalis and coronary sinus. However, angles in
cases with multiple pathways overlap. Yet, last 50% map helps
in separating the combined pathways from the solitary Bachmann
bundle pathway. Allocation to three different types, based on the
two magnetocardiographic map orientations, permits identification of conduction pathways in 27 of the 29 cases.
The prolongation and morphological changes of P-wave have been
related to inter-atrial block and propensity to AF.12,17,18 In addition
to inter-atrial conduction impairment, conduction barriers within
the LA have been shown.7,8 It is also possible that collision of electrical impulses through different propagation routes may be linked
to the pathogenesis of AF. Recognizing the inter-atrial conduction
pathway may assist in identifying subgroups for patient-tailored
therapy.
Catheter ablation of the RA septal region36 and CS connections37 or trans-section of the anterior LA38 have been effective
in the treatment of AF in some patients. Ablation of CS and FO
areas has altered inducibility to AF in an animal model.39 Thus,
all the three inter-atrial conduction pathways seem to have relevance in generation of AF, and knowledge on atrial conduction
pathways may have impact in refining methods for the ablation
treatment in patients with paroxysmal AF.
Limitations
Table 3 The relationship between left atrial
breakthrough site in electroanatomic mapping and type
of magnetic field orientation over the early (40 –70 ms)
and later (last 50%) part of left atrial depolarization
analysed from magnetocardiographic recordings
Breakthrough site to LA
MCG map type
.........................................
Type 1
Type 2
Type 3
................................................................................
BB (n = 14)
14
0
0
FO (n = 3)
0
3
0
CS (n = 2)
Combined routes (n = 10)
0
2
0
8
2
0
BB + FO
1
2
0
BB + CS
FO + CS
1
0
0
4
0
0
0
2
0
16
11
2
BB + FO + CS
All (No = 29)
The figures represent the number of cases in breakthrough site groups. BB,
Bachmann bundle; FO, rim of fossa ovalis; Combined, two or more breakthrough
areas; CS, coronary sinus ostium; LA, left atrium; MCG map, magnetocardiographic
map. The MCG map types refer to classification of cases based on pseudocurrent
orientation during left atrial activation: Type 1, pseudocurrent angle is positive in
both maps; Type 2, pseudocurrent angle is positive in the map over 40–70 ms
and negative in the map over last 50%; Type 3, pseudocurrent angle is negative in
both maps.
Correspondence of electroanatomic and MCG atrial mappings was
examined in patients with relatively normal hearts and highly symptomatic paroxysmal AF. Conductive properties might be different
in healthy subjects and when AF is associated with heart diseases.
The time windows for MCG map analysis may need adjustment to
cover the intended atrial compartments in markedly enlarged atria.
The influence of scars to MCG maps could not be evaluated in this
patient series mostly without structural heart disease.
The local representativeness of recorded intra-cardiac signals is
crucial for construction of the maps, especially around the postulated pathways. Registering potentials can be technically challenging
in the inter-atrial septum due to far-field potentials from nearby
structures. The number of patients in conduction pathway subgroups was small and therefore the performance of the technique
needs to be tested in larger populations. Due to necessity to rely
on visually observed direction of signal propagation in electroanatomic mapping, the authenticity of direction of atrial activation
determined by MCG mapping could not be ensured. Yet, this information is not necessary for validation of the ability to identify
inter-atrial conduction pathways.
Conclusions
Magnetocardiographic mapping over the frontal chest and subsequent analysis applying pseudocurrent distribution are capable
of identifying different activation breakthrough sites in the LA
during sinus rhythm with an adequate accuracy. This non-invasive
technique may also be used for assessing inter-atrial conduction
176
R. Jurkko et al.
in large patient series and healthy subjects, in which the invasive
measurements are not possible. Recognizing the inter-atrial conduction pathways may assist in understanding the pathogenesis of
AF and identifying subgroups for patient-tailored therapy.
Conflict of interest: none declared.
Funding
This study was supported by grants from the Finnish Foundation for
Cardiovascular Research, Helsinki, Finland and Alfred Kordelin Foundations, Helsinki, Finland.
Appendix
Interpolation with magnetic
multipole expansion
For calculation of the magnetic field map orientation and visualization of the measured field, the magnetic field was interpolated
from the measured data using magnetic multipole expansion.28
Magnetic multipole expansion is a series expansion of the magnetic
scalar potential, analogous to the electric multipole expansion.
Static magnetic field can be expressed by a scalar potential. In
bioelectromagnetic fields, the rate of change is relatively low,
and generally utilized the assumption of static fields is valid.
Then, in source-free region, such as the sensor surface, the curl
of the magnetic flux vanishes (r B~ ¼ 0) and the magnetic field
can be represented as the gradient of the magnetic scalar potential
~ ¼ rVm ). Because magnetic field has no sources (r H
~ ¼ 0),
(H
the magnetic scalar potential obeys the Laplace equation
(r2 Vm ¼ 0). We use truncated general serial solution of the
Laplace equation in spherical co-ordinates28
Vm ¼
5 X
l
X
Alm r l Ylm ðq; wÞ;
l¼1 m¼l
where Alm are coefficients for the sources inside a spherical surface
and Ylm the spherical harmonic functions. We set the expansion
origin at 15 cm below the centre of the sensor area (approximately
~ ¼ rVm and the sensor geometry for all
at the heart). Using H
sensors, we get a set of equations for coefficients Alm. These
equations can be expressed in matrix form by defining transfer
matrix T and the corresponding vector ~x of the multipole coefficients Alm. Then the coefficients are found from the measured
signals b~m by ~x ¼ T 1 b~m :
The interpolation is based on virtual sensors and composing
transfer matrix T0 for them, as explained above. The interpolated
signals are then obtained as b~i ¼ T 0 T 1 b~m :
For magnetic field interpolation, a square grid of points was generated at the sensor array surface with grid constant of 0.5 cm and
diameter of 26 cm. The field component Bz perpendicular to the
surface was interpolated at each time instant by assuming a
virtual magnetometer at each grid point. Surface gradients @(Bz)/
@x and @(Bz)/@y were calculated as the difference of adjacent
point values of interpolated Bz data divided by the 0.5 cm
separation.
Determining and
parameterization of the magnetic
field map orientation
In this work, the MCG maps were visualized and parameterized
using pseudocurrent transformation, originally presented by
Cohen and Hosaka.29 Pseudocurrents are 908 rotated magnetic
field surface gradient vectors, a~ ¼ @ðBz Þ=@y^ex @ðBz Þ=@x^ey ; that
reflect the underlying source currents. We used pseudocurrent
direction to determine the MCG map orientation.
For each MCG map, the distribution of pseudocurrent magnitude and direction were computed, with zero angle direction
pointing from subject’s right to left and positive clockwise. In
order to produce a robust measure of MCG map orientation
and to visually assess its significance, the pseudocurrents with relative strength above 70% in each map were selected for further
analysis. The orientation of the MCG map was defined as the
mean direction of the selected pseudocurrents.
References
1. O’Donnell D, Bourke JP, Furniss SS. Interatrial transseptal electrical conduction:
comparison of patients with atrial fibrillation and normal controls. J Cardiovasc
Electrophysiol 2002;13:1111 –7.
2. Oral H, Ozaydin M, Chugh A, Scharf C, Tada H, Hall B et al. Role of coronary
sinus in maintenance of atrial fibrillation. J Cardiovasc Electrophysiol 2003;14:
1329 –36.
3. Hindricks G, Kottkamp H. Simultaneous noncontact mapping of left atrium in
patients with paroxysmal atrial fibrillation. Circulation 2001;104:297 –303.
4. De Ponti R, Ho SY, Salerno-Uriarte JA, Tritto M, Spadacini G. Electroanatomic
analysis of sinus impulse propagation in normal human atria. J Cardiovasc Electrophysiol 2002;13:1 –10.
5. Lemery R, Soucie L, Martin B, Tanq AS, Green M, Healey J. Human study of biatrial electrical coupling: determinants of endocardial septal activation and conduction over interatrial connections. Circulation 2004;110:2083 –9.
6. Lemery R, Birnie D, Tang AS, Green M, Gollob M, Hendry M et al. Normal atrial
activation and voltage during sinus rhythm in the human heart: an endocardial and
epicardial mapping study in patients with a history of atrial fibrillation. J Cardiovasc
Electrophysiol 2007;18:402 – 8.
7. Markides V, Schilling RJ, Ho SY, Chow AW, Davies DW, Peters NS. Characterization left atrial activation in the intact human heart. Circulation 2003;107:733 –9.
8. Betts TR, Roberts PR, Morgan JM. High-density mapping of left atrial endocardial
activation during sinus rhythm and coronary sinus pacing in patients with paroxysmal atrial fibrillation. J Cardiovasc Electrophysiol 2004;15:1111 –7.
9. Holmqvist F, Husser D, Tapanainen JM, Carlson J, Jurkko R, Xia Y et al. Interatrial
conduction can be accurately determined using standard 12-lead electrocardiography: validation of P-wave morphology using electroanatomic mapping in
man. Heart Rhythm 2008;5:413 –8.
10. Roithinger FX, Cheng J, SippensGroenewegen A, Lee RJ, Saxon LA,
Scheinman MM et al. Use of electroanatomic mapping to delineate transseptal
atrial conduction in humans. Circulation 1999;100:1791 –7.
11. Cosio FG, Martin-Penato A, Pastor A, Núnez A, Montero MA, Cantle CP et al.
Atrial activation mapping in sinus rhythm in the clinical electrophysiology laboratory: observations during Bachmann’s bundle block. J Cardiovasc Electrophysiol
2004;15:524 –31.
12. Bayés de Luna A, Cladellas M, Oter R, Torner P, Guindo J, Marti V et al. Interatrial
conduction block and retrograde activation of the left atrium and paroxysmal
supraventricular tachyarrhythmia. Eur Heart J 1988;9:1112 –8.
13. Bachmann G. The inter-auricular time interval. Am J Physiol 1916;41:309 –20.
14. Wang K, Ho SY, Gibson DG, Anderson RH. Architecture of atrial musculature in
humans. Br Heart J 1995;73:559–65.
15. 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 –52.
16. Platonov PG, Mitrofanova LB, Chireikin LV, Olsson SB. Morphology of interatrial
conduction routes in patients with atrial fibrillation. Europace 2002;4:183 –92.
17. Ariyarajah V, Asad N, Tandar A, Spodick DH. Interatrial block: pandemic prevalence, significance and diagnosis. Chest 2005;128:970 –5.
MCG mapping of conduction to LA
18. Platonov PG, Carlson J, Ingemansson MP, Roijer A, Hansson A, Chireikin LV et al.
Detection of interatrial conduction defects with unfiltered signal-averaged P wave
ECG in patients with lone atrial fibrillation. Europace 2000;2:32–41.
19. Siltanen P. Magnetocardiography. In Macfarlane P, Lawrie TDV (eds). Comprehensive Electrocardiology Volume II. Oxford: Pergamon Press 1989 pp. 1405 –38.
20. Korhonen P, Montonen J, Endt P, Mäkijärvi M, Trahms L, Katila T et al. Magnetocardiographic intra-QRS fragmentation analysis in the identification of patients
with sustained ventricular tachycardia after myocardial infarction. Pacing Clin
Electrophysiol 2001;24:1179 –86.
21. Fenici R, Nenonen J, Pesola K, Korhonen P, Lötjönen J, Mäkijärvi M et al.
Nonfluroscopic localization of an amagnetic stimulation catheter by multichannel
magnetocardiography. Pacing Clin Electrophysiol 1999;22:1210 –20.
22. Mäntynen V, Vitikainen A-M, Koskinen R, Mäkijärvi M, Toivonen L, Montonen J.
Magnetocardiography is sensitive to differences in inter-atrial conduction in
patients with paroxysmal lone atrial fibrillation. In Cheyne D, Ross B,
Stroink G, Weinberg H (eds). Proceedings of the 15th International Conference on
Biomagnetism. Vancouver, BC, Canada: 2007 pp. 508 –11.
23. Koskinen R, Väänänen H, Mäntynen V, Montonen J, Nenonen J, Lehtonen L et al.
Field heterogeneity in magnetocardiographic atrial signals in patients with focallytriggered lone atrial fibrillation (Abstract). Heart Rhythm 2004;1:226.
24. Smeets JL, Ben-Haim SA, Rodriquez LM, Timmermans C, Wellens HJ. New
method for nonfluoroscopic endocardial mapping in humans. Circulation 1998;
97:2426 –32.
25. Gepstein L, Hayam G, Shlomo A, Ben-Haim SA. A novel method for nonfluoroscopic catheter-based electroanatomical mapping of the heart. Circulation 1997;
95:1611 –22.
26. Weiss C, Willems S, Rueppel R, Hoffmann M, Meinertz T. Electroanatomic
mapping (CARTO) of ectopic atrial tachycardia: impact of bipolar and unipolar
local electrogram annotation for localization the focal origin. J Interv Card Electrophysiol 2001;5:101 –7.
27. Koskinen R, Lehto M, Väänänen H, Rantonen J, Voipio-Pulkki L-M, Mäkijärvi M
et al. Measurement and reproducibility of magnetocardiographic filtered atrial
signal in patients with lone atrial fibrillation and in healthy subjects.
J Electrocardiol 2005;38:330 –6.
177
28. Jackson JD. Classical Electrodynamics. 3rd ed. New York, NY: John Wiley & Sons,
Inc.; 1999.
29. Cohen D, Hosaka H. Part II—Magnetic field produced by a current dipole.
J Electrocardiol 1976;9:409 –17.
30. Zar JH. Biostatistical Analysis.. 4th ed. Upper Saddle River, NJ: Prentice-Hall; 1999.
p630, 649.
31. Kim D, Kim K, Lee YH, Ahn H. Detection of atrial arrhythmia in superconducting
quantum interference device magnetocardiography; preliminary result of a
totally-non-invasive localization method for atrial current mapping. Interact Cardiovasc Thorac Surg 2007;6:274 –9.
32. Nakai K, Kawazoe K, Izumoto H, Tsuboi J, Oshima Y, Oka T et al. Construction of
a three-dimensional outline of the heart and conduction pathway by means of a
64 channel magnetocardiogram in patients with atrial flutter and fibrillation. Int J
Cardiovasc Imaging 2005;21:555 – 61.
33. Haberkorn W, Steinhoff U, Burghoff M, Kosch O, Morguet A, Koch H. Pseudo
current density maps of electrophysiological heart, nerve or brain function and
their physical basis. Biomagn Res Technol 2006;4:5.
34. Edenbrandt J, Pahlm O. Vectorcardiogram synthesized from 12-lead ECG: superiority of the inverse Dower matrix. J Electrocardiol 1988;21:361–7.
35. Carlson J, Havmöller R, Herreros A, Platonov P, Olsson B. Can orthogonal lead
indicators of propensity to atrial fibrillation be accurately assessed from the
12-lead ECG? Europace 2005;7:39 –48.
36. Gaita F, Riccardi R, Calò L, Scaglione M, Garberoglio L, Antolini R et al. Atrial
mapping and radiofrequency catheter ablation in patients with idiopathic atrial
fibrillation. Electrophysiological findings and ablation results. Circulation 1998;97:
2136 –45.
37. Haissaguerre M, Hocini M, Takahashi Y, O’Neill MD, Pernat A, Sanders P et al.
Impact of catheter ablation of the coronary sinus on paroxysmal or persistent
atrial fibrillation. J Cardiovasc Electrophysiol 2007;18:378 –86.
38. Sanders P, Jais P, Hocini M, Hsu LF, Scavée C, Sacher F et al. Electrophysiologic
and clinical consequences of linear catheter ablation to transect the anterior
left atrium in patients with atrial fibrillation. Heart Rhythm 2004;1:176 – 84.
39. Ott P, Kirk MM, Koo C, He DS, Bhattacharya B, Buxton A. Coronary sinus and
fossa ovalis ablation: effect on interatrial conduction and atrial fibrillation.
J Cardiovasc Electrophysiol 2007;18:310 –7.