1075
Spontaneous Tachyarrhythmias After
Cholinergic Suppression in the Isolated
Perfused Canine Right Atrium
Richard B. Schuessler, Leonid V. Rosenshtraukh, John P. Boineau,
Burt I. Bromberg, and James L. Cox
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
Atrial fibrillation occurs spontaneously after bradycardia induced by acetylcholine infusion or
vagal stimulation. To determine the mechanism of initiation of this tachyarrhythmia, we
infused acetylcholine (5 ml, 10` M) into Krebs-Henseleit-perfused isolated canine right atria
(n=10). Unipolar electrograms were recorded from 250 sites simultaneously during control
rhythms, pacing (cycle length=300 msec) with and without acetylcholine, and recovery of
spontaneous activity. Activation sequence maps were constructed from each recording. Stable
spontaneous rhythm was present in all preparations during control conditions. Activation
sequence maps, recorded during continuous pacing with and without acetylcholine, demonstrated no dromotropic changes due to the acetylcholine. Focal asynchronous recovery of
spontaneous activity was initiated from different sites, resulting in bigeminal or trigeminal
premature depolarizations in 41 of 73 cases after infusion of acetylcholine. A reentrant
tachyarrhythmia was initiated in 24 of 41 cases by the closely coupled recovery beats
(A,A2=100+37 msec; A2A3=97+27 msec). The reentry was initiated by interaction of the
premature impulse with regions of functional block that were a result of the cholinergically
induced dispersion of refractoriness. All the tachyarrhythmias terminated spontaneously, and
stable spontaneous control rhythms returned. In conclusion, the data suggest that the
premature depolarizations that initiate the reentrant tachyarrhythmia are caused by the
asynchronous recovery of multiple right atrial pacemakers accompanied by variable entrance
block at the later depolarizing sites. (Circulation Research 1991;69:1075-1087)
It long has been recognized that vagal stimulation
can initiate atrial fibrillation.1 In dogs, vagal
stimulation or injection of a cholinergic agonist
into the sinus node artery depresses spontaneous
activity and often is followed by the onset of atrial
fibrillation.23 The mechanisms of initiation are not
clearly understood. Allessie et a14 have shown that,
once initiated, cholinergic atrial fibrillation in the dog
From the Division of Cardiothoracic Surgery (R.B.S., J.P.B.,
J.L.C.), Department of Surgery, Washington University School of
Medicine, Barnes Hospital, St. Louis, Mo.; the Institute of Experimental Cardiology (L.V.R.), USSR Cardiology Research Center,
Moscow, USSR; and the Division of Pediatrics/Adolescent Medicine (B.I.B.), St. Louis University Medical Center, St. Louis, Mo.
Supported by grants R01 HL-32257 and R01 HL-33722 from the
National Institutes of Health, and grant 88 1223 from the American Heart Association (AHA). Funds contributed in part by the
AHA, Missouri Affiliate, and by the Joint US/USSR Health
Exchange in the Cardiopulmonary Area, Program Area 5 - Sud-
den Death.
Address for correspondence: Richard B. Schuessler, PhD, Division of Cardiothoracic Surgery, Box 8109 - 3308 CSRB, Washington University School of Medicine, 660 S. Euclid Ave., St.
Louis, MO 63110.
Received December 13, 1990; accepted June 7, 1991.
is a reentrant arrhythmia. However, in that study, the
tachyarrhythmia was initiated artificially using the
extrastimulus technique. Two conditions are required to initiate a reentrant tachyarrhythmia. First,
there must be an intrinsic substrate that causes
nonuniform conduction and unidirectional block.
Second, a triggering event is needed, such as a
premature depolarization with a short coupling interval or complex train of closely coupled depolarizations. These are simulated in experimental and clinical electrophysiological studies by programmed
extrastimulation. However, the spontaneous triggering event has not been described in mammalian
tissue. Nadeau et a15 have suggested that, after
cholinergic suppression, the sinus node and supraatrioventricular nodal junctional pacemakers participate in the initiation of the tachyarrhythmia. Several
studies6-9 have demonstrated that, in addition to the
sinus node, the right atrium is populated with multiple pacemakers. These pacemakers, which can function under normal physiological conditions, have
been shown to be under autonomic control with a
differentiated responsiveness to nerve stimulation
1076
Circulation Research Vol 69, No 4 October 1991
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
and autonomic neurotransmitters. In addition to
decreasing automaticity, cholinergic agonists shorten
action potential duration and refractory period of
atrial myocytes. Most studies have suggested that
propagation velocity in normal atrial myocardium is
unaffected by cholinergic agonists.10 11 However,
propagation into and out of the sinus node is slowed
by acetylcholine (Ach) and there is an increased
incidence of exit and entrance block.1213 In a study by
Bonke et al,12 the data suggest that Ach can make the
sinus node inexcitable to direct stimulation. Rosenshtraukh et al14 have shown that vagal stimulation
produces regions of transiently inexcitable atrial myocardium in the frog. This transient inexcitability,
coupled with a single spontaneous sinus depolarization, initiated a reentrant tachyarrhythmia.
The objective of the present study was to map the
activation sequence of the isolated perfused canine
right atrium during the initiation of atrial tachyarrhythmias that occur spontaneously after the administration of Ach. The patterns of activation are
described by high-resolution activation sequence
mapping and give insights into the underlying mechanisms involved in the initiation of this arrhythmia.
The mapping demonstrates that the closely coupled
initial beats of the tachyarrhythmia originate focally,
followed by a transition to a reentrant pattern.
Materials and Methods
Normal mongrel dogs (n=15) weighing 20-25 kg
were intravenously anesthetized with 30 mg/kg of
pentobarbital sodium. The dogs were intubated and
placed on a positive pressure respirator (Harvard
Apparatus, South Natick, Mass.). A median sternotomy was performed, and the heart was cradled in the
pericardium. The azygous vein was ligated and divided. The interatrial groove was dissected, and the
right coronary artery was dissected free from its
origin to a point distal to the sinus node artery. The
dogs were heparinized with 100 gug/kg i.v. heparin.
The ventricular branches of the right coronary artery
were ligated. The right atrium was excised rapidly by
ligating and dividing the inferior and superior venae
cavae, cross clamping the aorta, infusing cold cardioplegia solution into the aortic root, and applying
iced saline topically to the heart. The proximal right
coronary artery was cannulated with polyethylene
tubing (PE-50 Intramedic, 0.58-mm i.d., 0.965-mm
o.d.). The ventricles and excess atrial tissue were
trimmed, and the atrium was unfolded and mounted
on a flat platform electrode. The mounted preparation was approximately 4.5x7.5 cm. The electrode
template consisted of 250 silver electrodes with an
interelectrode distance of 3-5 mm. The atrium then
was placed in a temperature-controlled bath at 37°C,
and the preparation was perfused with Krebs-Henseleit solution at a rate of 8-10 ml/min. The millimolar
composition of the solution was Na 143, K 4.7, Cl 129,
Ca 1.25, Mg 1.20, HCO3 25, and dextrose 11.1. At the
same time, the preparation was superfused continuously with the same solution. A bipolar pacing elec-
TABLE 1. Summary of Return Rhythms
Dog
1
2
3
4
5
6
7
8
9
10
Total (n)
Percent
No. of
infusions
4
5
2
2
10
10
10
10
10
10
73
Retumn rhythm
Bigeminal or
Single
trigeminal
Tachyarrhythmia
2
0
0
1
8
7
4
1
3
6
32
44
2
4
1
0
0
0
2
1
3
4
17
23
0
1
1
1
2
3
4
8
4
0
24
33
trode was sutured to the preparation. Pacing was
performed from the inferior aspect of the atrium to
avoid stimulating nerve fibers in the superior portion
of the preparation.15 Spontaneous activity resumed in
all the preparations within 5 minutes of the initiation
of the initial perfusion. After the start of spontaneous activity, a period of 30 minutes was allowed for
stabilization of the preparation before any data were
taken. Further details of the technique and the
normal electrophysiology of the preparation have
been reported previously."
Control unipolar electrograms were recorded from
all 250 sites simultaneously at the start of each study.
Data also were recorded during pacing at a cycle
length of 300 msec. Ach at a dose of 10-3'5 M was
infused continuously for 3 minutes. This dose was
above the chronotropic Em., previously determined
for this preparation and always produced asystole for
the period of the infusion." Pacing data (cycle
length=300 msec) were recorded during the infusion.
Then for each preparation, an average of seven bolus
doses (5 ml) of 10-3-5 M Ach were given, which
produced a 232±35-second period of arrest. The
return of spontaneous activity was recorded. Ten
minutes were allowed between each Ach infusion for
the preparation to recover. This protocol was followed in 10 preparations. The number of infusions
given in each preparation is shown in Table 1.
Because preliminary observations suggested that
the high dose of Ach had little effect on propagation
velocity and no areas of inexcitability were observed
similar to those observed in the frog,14 an additional
three preparations were studied in which the potassium levels and temperatures were reduced to values
normal for the frog (1.8 mM potassium, 21°C). The
above drug and pacing protocols were repeated. The
temperature then was raised to 37°C, and the drug and
pacing protocols were repeated. Two additional preparations were studied to determine if Ach caused the
release of endogenous norepinephrine. Five bolus
doses of Ach were administered, and data were re-
Schuessler et al Spontaneous Cholinergic Tachyarrhythmia
1077
sv-c
FIGURE 1. Computer activation time map is shown on a
schematic anatomic drawing. Each number represents the
activation time at the corresponding electrode site. Times
all are adjusted to the earliest time, designated as zero.
Computer-generated isochrones are shown every 10 msec.
Major anatomic landmarks are labeled. The preparation is
viewed from the epicardial perspective. SVC, superior vena
cava; IVC, infenior vena cava; ICB, intracaval band; CT,
crista terminalis; RAA, right atrial appendage; SNA, sinus
node artery.
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
:- SNA
'vc
corded. Then the infusion was switched to KrebsHenseleit with i0-` M propranolol, and the bolus
doses of Ach and data recording were repeated.
A 256-channel computerized data acquisition and
analysis system was used to collect, process, and
display data. Unipolar electrograms were recorded at
a gain of 1,000, with a frequency response of 0.1-500
Hz. A silver electrode was placed in the bath to
provide the indifferent input for the unipolar electrodes. Each channel was digitized at 1,000 Hz with
12-bit resolution. Local activation times were determined from the time of the maximum negative derivative (-dV/dt) of the electrogram. All electrograms
were edited visually to verify accuracy of the computer-picked activation time, and these activation times
were displayed on a schematic diagram of the atrium
as activation time maps. Figure 1 shows an example
of the computer display of a spontaneous normal
sinus beat.
Cycle length data were compared using the Student's t test. All statistical calculations were performed using the SYSTAT statistical package. Data are
expressed as mean+SD, and a value of p<0.05 was
considered significant.
All animals received humane care in compliance
with the Principles of Laboratory Animal Care formulated by the National Society of Medical Research
and Guide for the Care and Use of Laboratory Animals
prepared by the National Academy of Sciences (NIH
publication 80-23, revised 1978). In addition, the
study protocol was approved by the Washington
University Animal Studies Committee.
Results
Figure 2 shows the effect of steady-state infusion of
i0"- M Ach on the paced activation sequence
(pacing cycle length=300 msec). During control conditions (panel A), the activation started on the
inferior portion of the crista terminalis at the site of
the pacing electrode and spread asymmetrically to
the remainder of the atrium, with the most rapid
propagation occurring along the crista. Panel B
shows the paced activation sequence in the same
preparation during Ach infusion. During the period
of infusion, there was no spontaneous activity. The
activation sequence was similar to that shown in
panel A, with no areas of conduction slowing or
regions of inexcitability. This example was typical of
all the pacing data that had little or no change in
propagation. When any change did occur, it usually
was a slight increase in the propagation velocity.
Continuous pacing at 300 msec during Ach infusion
did not induce extra beats. In the three preparations
in which the potassium (1.8 mM) and temperature
(21°C) were lowered, Ach produced no conduction
slowing or inexcitability. With low potassium and
normal temperature (37°C), Ach also had no effect
on propagation; however, the lower temperature did
reduce the propagation velocity and intrinsic rate.
Three patterns of return activity were observed
after Ach administration. Figure 3 shows example
electrograms. The first pattern (panel A) was a
gradual return to control rhythm, in which a single
return beat was followed by a second at a coupling
Circulation Research Vol 69, No 4 October 1991
1078
A.
B.
FIGURE 2. Activation sequence of a
typical beat during continuous pacing
(cycle length [CL] =300 msec) is
shown without (panel A) and with
(panel B) acetylcholine (Ach). The
asterisk marks the pacing site. ANT,
anterior; POST, posterior; SVC, superior vena cava; IVC, inferior vena cava;
ICB, intracaval band; CT, crista terninalis; RAA, tight atrial appendage;
SNA, sinus node artery.
120 mm
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
Ivc
Paced CL=300ms
Paced+IO15M Ach
CL=300ms
interval of greater than 1,000 msec. The cycle gradually decreased to the control cycle length. This occurred after 32 of 73 infusions. The second pattern,
observed in 17 of 73 infusions, was a bigeminal or
trigeminal rhythm (panels B and C) in which the
coupling intervals were <200 msec. In 24 of 73 infusions, the third pattern (panels D-F), a tachyarrhythmia, occurred. All the tachyarrhythmias terminated
spontaneously within 60 seconds. Closely coupled
return beats also occurred in the two preparations
with propranolol in the perfusate.
Figure 4 shows two examples of the activation
sequence associated with the bigeminal pattern of
response. In the first example (top two panels), the
activation started focally (A1) at a site on the inferior
portion of the preparation. The activation propagated with a normal velocity to the remainder of the
atrium. At 157 msec after A1, a second focal site
depolarized (A2) in the area of the intracaval band
and activated the remainder of the atrium. The next
beat occurred more than 600 msec later. No regions
of block or slowing of propagation were observed. In
another example (bottom two panels), activity started
in the sinus node region (A1) and spread to the
remainder of the atrium in a manner similar to the
control rhythm in this preparation. A second beat
(A2) started focally in the lateral portion of the
atrium from two sites at 99 and 100 msec after the
initiation of the A1 beat. The two activation wave
fronts fused and blocked along the medial aspect of
the crista terminalis. The wave fronts rotated around
the lines of block; however, the wave front did not
reenter, and the next beat occurred more than 600
msec later.
Figures 5-8 illustrate representative patterns of
initiation of spontaneous tachyarrhythmias. In the first
example, Figure 5, the first return beat (A1) was
multifocal, with activation starting at two separate
sites. The superior site depolarized first, and 3 msec
later, a site approximately 25 mm inferior to the first
site depolarized. The wave fronts from the two sites
fused, rapidly activating the entire atrium within 35
msec. At 72 msec, a second beat (A2) was initiated at
a second inferior site, and the activation wave front
spread out, blocking both inferiorly and superiorly.
One superior line of block prevented activation of the
superior lateral aspect of the atrium. The wave front
rotated around the inferior line of block in a clockwise
direction and reentered (A3) at 125 msec. Again, the
A3 wave front blocked inferiorly and superiorly at
several sites. A larger segment of the superior atrium
was unable to activate because of the increased block.
At 197 msec, the wave front reentered a different
inferior site medial to the first. The line of block on the
superior portion of the atrium then extended across
the whole preparation, thus preventing activation of
the superior one third of the preparation. Subsequent
beats were reentrant with a cycle length of 50-60
msec. Figure 6 illustrates example electrograms recorded during this tachyarrhythmia.
In another example, Figure 7, initial activation
started superiorly and activated the remainder of the
atrium in a normal fashion. In the second beat (A2),
two focal sites initiated activation at 75 and 80 msec.
The two beats fused, and a region of block developed
in the lateral inferior atrium. At 146 msec (A3),
another superior focal site initiated activation. The
site was lateral to the superior site that initiated A2.
However, at 162
msec,
the
wave
front that
was
Schuessler et al Spontaneous Cholinergic Tachyarrhythmia
A
O
1.00
E -1.00
-300
-o00
ms
0
200
400
600
800
1000
0
200
400
600
800
1000
20
4
0
200
400
600
800
1000
0
200
400
600
800
1000
-1.00
-3.00
-5.00
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
1.00
-3
0
0
1
FIGURE 3. Examples ofunipolar electrograms shown for the
initial return rhythm after six separate episodes of acetylcholine-induced arrest in the same preparation (dog 7). Electrograms all are recorded from the same site. Panel A: Single
return beat; panel B: bigeminal pattem; panel C: trigeminal
pattem. In panels A-C, the initial beats all are followed by a
long diastolic period, with a gradual return to the control rate.
Panels D-F: Three different spontaneous tachyarrhythmias.
initiated at 75 msec and that rotated around the line
of block reentered in the inferior lateral atrium (A3).
The wave front initiated by the focal site of activation
at 146 msec and the wave front initiated by reentry at
162 msec collided with a longer line of block developing in the inferior lateral portion of the preparation. The wave front rotated around the line of block,
reentering at 197 msec (A4), forming a figure eight
reentry pattern. Until the tachyarrhythmia terminated, this pattern dominated, with the reentry occurring every 40-50 msec.
Figure 8 shows the first four beats of another
spontaneous tachyarrhythmia. In the first beat, A1,
the initial site of activation was medial to the crista
terminalis. The second beat, A2, started focally 106
msec after A1. Two regions of block developed, one
in the superior portion of the preparation paralleling
the medial portion of the crista terminalis, and the
second in the inferior portion of the atrium. None of
the wave fronts rotating around two lines of block
reentered. At 213 msec, an activation started focally
1079
at the junction of the inferior vena cava and atrium
(A3). The wave front blocked in two directions and
rotated around the lines of block, forming a figure
eight pattern. The medial limb of the circuit reentered at 272 msec, initiating A4, and the pattern
repeated. The superior portion of the preparation
was unable to activate with each beat and alternately
activated every other beat.
Figures 5-8 represent the patterns of initiation in
all but one of the spontaneous tachyarrhythmias. In
Figure 5, the reentry was initiated by a bigeminal
focal pattern of activation initiation. Figure 8 shows a
trigeminal pattern initiating a reentrant pattern. Figure 7 illustrates a trigeminal pattern in which the
second focal pattern of initiation initiated the reentry
and fused with the third focally initiated wave front.
With the exception of one tachyarrhythmia, when a
bigeminal or trigeminal focal return pattern occurred, the successive focal beats never originated
from the same site. The one tachyarrhythmia pattern
that was different consisted of four focally activated
beats with a cycle length of 210 msec. The site was
located in the inferior lateral portion of the atrium. It
was the only preparation in which the initial closely
coupled return beats were initiated from the same
site.
A summary of all 10 animals is given in Table 1. Eight
of 10 of the preparations had a spontaneous tachyarrhythmia. When repeated tachyarrhythmias occurred
in the same preparations, the patterns of initiation were
not the same (Figure 3, panels D-F). The mean (±SD)
coupling interval of the bigeminal focal return pattern
that initiated a reentrant tachyarrhythmia was 94±27
msec, and those that did not had a mean of 123 ±23
(p=0.01). In the four occurrences of trigeminal focal
initiation patterns, the coupling intervals were
A1A2=129-56 msec and A2A3=86+27 msec. The sites
of activation initiation of all the control, bigeminal, and
trigeminal beats are plotted on a schematic outline of
the atrium in Figure 9. The greatest density of sites
parallels the crista terminalis.
Discussion
Several studies have suggested that cholinergically
mediated atrial tachyarrhythmias are reentrant arrhythmias.414 The underlying mechanism for unidirectional block, essential for the initiation of re-ntry,
is probably nonuniform repolarization.4 Another essential mechanism for the initiation of reentry is a
triggering impulse. Several studies have used electrical or mechanical stimulation to induce the tachyarrhythmia.4,516 Most previous data have shown that
either a single or multiple premature coupled depolarization is required to induce tachyarrhythmias.
Few studies have examined the spontaneous mechanisms for the initiation of these types of atrial tachyarrhythmias. One such study by Rosenshtraukh et
al14 has shown that, in the frog, a single wave front
starting from the sinus venosus during vagal stimulation encounters a region made transiently inexcitable
by the vagal stimulation. The activation wave front
1080
Circulation Research Vol 69, No 4 October 1991
A1
A2
FIGURE 4. Two examples of
I20mm
bigeminal return rhythms.
Rhythm 1 (top panels) shows
activation starting in the lower
portion of the atrium int A, and
the lateral portion near the intraDownloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
2
A1
A2
caval band 157 msec later in A2.
In rhythm 2 (bottom panels),
activation starts in A1 at the medial side of the cista terminalis.
In A2, the activation starts multifocally in the lateral atrium. The
wave front blocks (thick dark
line) along the medial edge of the
crista terminalis. ANT, anterior;
POST, posterior; SVC, superior
vena cava; IVC, inferior vena
cava; ICB, intracaval band; CT,
crista terminalis; RAA, right
atrial appendage; SNA, sinus
node artery.
20 mm
blocks on these regions, which then dissipate, permitting the wave front, which has rotated around the
blocked region, to reenter. No premature coupled
depolarizations were needed in this study. In the
present study, unlike studies in amphibians, Ach did
not result in regions of block at long (300 msec)
coupling intervals. Pacing data (Figure 2) were consistent with previous studies in which Ach was shown
not to affect propagation velocity. Although some
studies have suggested that Ach can accelerate propagation in mammalian atrial tissue,17 no studies have
shown inexcitability to occur as a result of vagal
stimulation or Ach infusion. Activation block or
slowing also was absent in all the initial return
depolarizations in the present study (the A1 beats in
Figures 4-8). In addition, during the pacing protocols, no coupled premature depolarizations were
observed to occur, indicating that the absence of
obvious reentry resulted from conduction block in
either myocardial or pacemaker tissue under the
influence of Ach.
The initial sites of the return of activation (A1)
were widely distributed (Figure 9) but were concentrated along the crista terminalis from the most
inferior to the superior portion of the preparation.
This spanned a distance of 60 mm, substantially
larger than the sinus node. Previous studies have
demonstrated that multiple pacemakers exist in the
right atrium.6-8 Boineau et a19 have shown that there
is a widespread system of pacemakers, termed the
pacemaker complex, that has a similar distribution to
that shown in Figure 9. With Ach having no effect on
propagation, it is unlikely that the widespread distribution of sites of initial activation were due to
Schuessler et al Spontaneous Cholinergic Tachyarrhythmia
o
A, A2
A4
A3
¢
1081
.,
-44
0
0
100
200
300
P.
400
500
A2
A,
I20 mm
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
A3
complex conduction from a single site. Therefore, it
is likely that these multiple sites of earliest activation
are associated with widely distributed pacemakers.
Several possible mechanisms could determine
which pacemaker initiated the first propagated atrial
depolarization (A1). The most straightforward reason
is that the pacemaker that has the fastest intrinsic
rate initiated activation. However, the first pacemaker to assume dominance could do so even if its
intrinsic rate was slower than other pacemakers, if
the Ach were cleared faster at that pacemaker because of nonuniform distribution of acetylcholinesterase or because of a differential sensitivity to Ach.
This would explain why the control pacemaker site,
A4
FIGURE 5. First four beats ofa spontaneous
tachyarrhythmia (A1-A4). An electrogram recorded from the lower portion of the preparation is shown above, with each complex labeled corresponding to the mapped beat
below. The vertical line before the A, waveform on the electrogram marks time zero. The
activation starts multifocally (A,), then at 72
msec (A2) the second beat is initiated, which
blocks inferiorly and superiorly. The wave
front rotates around the inferior line of block
and reenters in A3. It reenters again in A4.
The crosshatched areas are regions of tissue
that do not activate because of the block of
the activation wave front. ANT, anterior;
POST, posterior; SVC, superior vena cava;
IVC, inferior vena cava; ICB, intracaval
band; CT, crista terminalis; RAA, right atrial
appendage; SNA, sinus node artery.
which had the fastest intrinsic rate during control
conditions, was not always the first site to initiate a
depolarization after suppression with Ach. Another
possibility is that the site that dominated the initial
depolarization may not have been the pacemaker
that depolarized first. Rozanski and Lipsius13 demonstrated that Ach can produce complete exit block
from subsidiary pacemakers in the canine right
atrium. A pacemaker may resume automaticity before any other pacemaker but fail to initiate a
depolarization because of exit block. Another mechanism that could result in a pacemaker assuming
dominance is postwithdrawal excitation.18 Pacemaker
tissue in the canine coronary sinus, when exposed to
1082
Circulation Research Vol 69, No 4 October 1991
v
X * k ~~~~~z
*
.,
,i
4z
gt z z mkSzwws*
.
C
In
cu
t:::::
:
:N::::
.or
.....
..
...
.J
/
t-{.
.
l,.
...
..,.....
.....
..........
CD
cu
.ss
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
<.
./.
S.....
.
.... ,..
.... t..
\.
.
D
.. EF ....
Z.
m
we
:--T
q4
K
CD
'4
.....
....
C>
In
.........
C> C>
In
In
.
Ao.I
<
OI
.
A
.. . .
..
1....
C.
....
cu0W
00
c.
Aaoolol
I
..
c;
.
..
tm m m m
O
am
I;c;;I
C
C
Cu
....
.
.
CD ...
CD c;
C2C
CDC> ~CDCC
::c;
:n en n c;
-qiCD
CID
O O O ;hO
0
0I
C;
A;Cu C; C;A;
c;,c
0
W
d4dd
0dddd4 eO I dd Od d d d d
00d4
<
0J
m
N
<:t
m
0
0)
W
d04 .i
LL
Schuessler et al Spontaneous Cholinergic Tachyarrhythmia
2
AI
A2
A3
1083
A4
0
0
MV
100
200
300
400
A,
500
A2
I 20mm
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
ICB -
A3
A4
FIGURE 7. First four beats of a tachyarrhythmia (A1-A4). The electrogram above the
maps is recorded from the superior portion of
the preparation. The vertical line corresponds
to time zero (A1). The activation starts in A,
near the medial edge of the crista terminalis.
The second beat starts multifocally and blocks
in the lateral inferior portion of the atrium. In
the third beat, the activation starts focally in
the superior part of the atrium at 146 msec,
and the wave front from A2 reenters in the
inferior lateral atrium at 162 msec. This sets
up a reentrant pattem in A4 and subsequent
beats. ANT, anterior; POST, posterior; SVC,
supenor vena cava; IVC, inferior vena cava;
ICB, intracaval band; CT, crista terninalis;
RAA, right atrial appendage; SNA, sinus node
artery.
Ach, undergoes a dramatic hyperpolarization. When
Ach is withdrawn, the hyperpolarized membrane
rapidly rebounds and hypopolarizes above the initial
resting membrane potential, resulting in a depolarization similar to triggered activity. It is unlikely that
this occurred in pacemakers in this study, because
when reentry subsequently occurred, the cycle
lengths were short (40-60 msec), suggesting that
there was still a considerable concentration of Ach
present that kept the membrane hyperpolarized.
However, in the one case in which four successive
beats were initiated focally from the same site, some
other mechanism, such as triggered activity or intramural microreentry, could have been responsible.
In all the preparations, the initiation of the second
recovery beat (A2) was also focal. In addition, it was
always at a site separate from the first site A1, when
the A2 occurred at a short coupling interval (<200
msec). Several possible mechanisms could account
for this pattern of activation. These sites could have
been other pacemaker sites that were not reset by the
previous depolarization (A1). The failure to reset
could be due to entrance block that prohibited the A1
wave front from depolarizing the pacemaker. It also
1084
Circulation Research Vol 69, No 4 October 1991
toA,
MV
0
-4
0
A,
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
A3
A2 A3 A4
kI{ A'
_1
|
|
400
800
1
I'
J
---
|
1200
1600
A2
A4
is possible that the pacemaker was refractory, because it had previously depolarized but did not
propagate to the remainder of the atrium because of
exit block. However, as the Ach cleared and the block
dissipated, the unreset pacemaker depolarized and
propagated to the remainder of the atrium. Another
possible mechanism could be some form of reentry in
the pacemaker tissue, in which the A1 depolarization
wave front generated an echo beat out of another
pacemaker. This is unlikely because pacing during
Ach infusion never resulted in obvious conduction
block, coupled depolarizations, or tachyarrhythmia.
Triggered activity also is an unlikely mechanism for
the A2 depolarization, because Ach has been shown
FIGURE 8. First four beats of a spontaneous
tachyarrhythmia (A1-A4). Beats Al, A2, and A3
start focally. Block occurs in A2, but no reentry
occurs. The focal activation pattem in A3 sets up
a figure eight reentrypattem, with the medial area
of the circuit reentering in A4. The electrogram
above the map is recorded from the upperportion
of the preparation. ANT, anterior; POST, posterior; SVC, superior vena cava; IVC, inferior vena
cava; ICB, intracaval band; CT, crista terninalis; RAA, right atrial appendage; SNA, sinus node
artery.
to inhibit triggered activity, and again, continuous
pacing in the presence of Ach never resulted in extra
beats. Very fine or intramural microreentry in atrial
muscle also is unlikely, because Ach has no effect on
conduction, and, because of the long period of arrest,
none of the tissue was refractory to the A1 depolarization. It also is unlikely that the short coupling
interval of the A2 was the result of some form of
Ach-induced release of endogenous norepinephrine,
because 13-blockade did not abolish the closely coupled return patterns.
Depending on the coupling interval of the A1A2
depolarization, the A2 wave front could slow and
block (Figure 4, rhythm 2, A2). If the wave front
Schuessler et al Spontaneous Cholinergic Tachyarrhythmia
a
.
,20mm
IC.
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
oControl
MA2
*AI
&A3
FIGURE 9. Geographic distribution of sites of focal impulse
initiation are shown for the control rhythm (open circles) and
the focal bigeminal and trigeminal rhythms that initiated a
tachycardia. The first recovery beat A, (closed circles), second
recovery A2 (closed squares), and third focal recovery sites A3
(closed triangles) are plotted on a schematic of the atrial
anatomy. ANT, anterior; POST, posterior; SVC, superior vena
cava; IVC, inferior vena cava; ICB, intracaval band; CT,
crista terminalis; RAA, right atral appendage; SNA, sinus
node artery.
rotated around this region of block, it could reenter,
starting the reentrant phase of the tachyarrhythmia.
Figure 5 illustrates such an example, in which complex lines of block developed as a result of the A2
depolarization. The wave front was able to rotate
around the inferior region of block and reenter (A3).
Successive beats (A4... .) were reentrant. As the Ach
cleared, the tachyarrhythmia terminated within a few
seconds.
In Figure 7, a focal A1A2 pattern initiated a
reentrant circuit A3, A4,.... However, in the A3
depolarization, two sites initiated the beat, a superior
focal site at 146 msec and an inferior lateral reentrant circuit at 162 msec. The two wave fronts fused,
and the reentrant circuit dominated the activation in
successive beats. The mechanism for the A3 superior
focal depolarization could be either automatic or
reentrant because of a small intramural microreentrant circuit or because of a dissociation of epicardial
and endocardial activation, resulting in a threedimensional circuit. This mechanism has been suggested by Zaitsev et al,19 who mapped a canine atrial
epicardium and endocardium simultaneously with a
limited number of electrodes during Ach-dependent
tachyarrhythmia.
In four tachyarrhythmias, the A2 depolarization
did not reenter, and
a
focal pattern of activation
1085
initiated the A3 depolarization. Figure 8 illustrates
such a pattern, in which the A2 wave front did not
appear to reenter, despite the fact that several regions of block occurred. However, a focal depolarization started at the junction of the inferior vena
cava and atrium that did block and reenter. This
lower site was in the region of a well-defined subsidiary pacemaker characterized by Rozanski and Lipsius.13 An obvious reentrant circuit was not mapped,
and it is unlikely that this focal site of initiation was
due to intramural microreentry, as the tissue was
extremely thin in this region (<1 mm).
The multifocal asynchronous return of activity not
only initiated successive beats, but a single beat also
could be initiated by a multifocal pattern of activation. Examples of this pattern are shown in Figure 4,
rhythm 2, A2 and Figures 5 A1 and 6 A2. In each
example, the depolarization wave front was initiated
from two sites within 5 msec of each other. The
asynchronous return of focal activity suggests that if
the underlying mechanisms for the focal activity were
automaticity, then the recovery times for these pacemakers would be different after cholinergic suppression. If they returned at closely coupled intervals
(0-10 msec, relative to the initial return), the result
would be a multicentric origin pattern of a single
beat. If they were to return at an interval longer than
that (10-50 msec), it is likely that the spreading
depolarization wave front would render the surrounding atrial tissue refractory and the later pacemaker site would not participate in the initiation of
the beat. If the return interval were greater than the
refractory period (>50 msec), then a wave front
could be initiated. This wave front, now a "premature" depolarization, then could potentially initiate a
reentrant wave front.
The second wave front A2 always was initiated
focally but then often encountered regions of slow
conduction and functional block. The shorter the
coupling interval, the higher the probability was that
these lines or regions of block occurred (Figure 4). In
all the cases in which a tachyarrhythmia of greater
than four beats was initiated, the A2 or A3 wave front
reentered around a line of block, giving rise to the
successive beats. The lines of block were unstable
and often changed their shape and location. The
mechanism of block was probably a cholinergically
induced refractory period inhomogeneity. However,
sometimes block occurred along the edge of the
crista terminalis (Figure 4, rhythm 2, A2 and Figure 8
A2), suggesting that tissue anisotropy could be the
underlying mechanism of block. But the data obtained in the present study do not differentiate
between the two mechanisms. As the Ach cleared,
the tachyarrhythmia always terminated spontaneously. The interval between successive beats of the
tachyarrhythmia could prolong (electrogram, Figure
8) after the reentry started or could terminate
abruptly as a result of the prolongation of the refractory period as the Ach cleared.
1086
Circulation Research Vol 69, No 4 October 1991
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
If asynchronous depolarization were the underlying mechanism generating premature atrial depolarizations, then within a critical time interval three
conditions would have to be met. The first condition
requires that multiple functioning pacemakers must
exist in the atrium. Studies in both dogs and humans
have demonstrated that multiple pacemakers exist in
the atria.6-9,20 In these studies, there appear to be
different rate ranges and sensitivity to neurotransmitters for each of these pacemakers. Under normal
conditions, most of these pacemakers are suppressed
by the spread of depolarization wave fronts. The
second condition is entrance block into one or more
of the pacemaker sites at the time when spontaneous
activity returns. Studies by Bonke et a121 have demonstrated that during atrial fibrillation, a sinus node
pacemaker can depolarize spontaneously, independent of fibrillatory activity in the surrounding atrial
tissue. Ach also has been shown to affect conduction
into and out of the sinus node.'213 Although Ach is
not present physiologically in the concentrations
used in this study, some vagal tone almost always is
present. Clinically, Coumel22 has shown that patients
with high vagal tone have spontaneous atrial fibrillation associated with increased vagal activity. The
third condition for a premature impulse to be generated requires that the time of depolarization of the
underlying pacemaker must occur when the surrounding atrial tissue is excitable and any exit block
out of that pacemaker has dissipated. The fact that
the patterns of initial activation were not the same
within any one preparation suggests that the conditions necessary for an asynchronous depolarization
sequence to occur are very sensitive to the underlying
metabolic state of the preparation, as well as to the
precise amount and timing of Ach delivery. For
reentry to occur as a result of the premature depolarization, a substrate for unidirectional block must
be present, and the premature depolarization must
occur within that critical interval when reentry is
possible. It is interesting to note that studies of the
coupling interval of premature atrial contraction in
patients with spontaneous atrial fibrillation show a
similar pattern to those demonstrated in the present
study.23'24 In those studies and the present study, the
premature beats that occurred were more likely to
initiate fibrillation at the shorter coupling intervals.
In conclusion, detailed mapping of the return of
spontaneous activity after suppression with Ach in
the isolated perfused canine right atrium has demonstrated that closely coupled return beats originated
focally from different sites. Furthermore, if the coupling interval was within a critical time interval, a
reentrant tachyarrhythmia could be initiated. It is
suggested that the focal sites are pacemakers that
recover spontaneous activity asynchronously. If one
or more sites were protected by entrance block,
single or double "premature" depolarizations could
be generated. Although the conditions in this study
were nonphysiological, the initiation patterns of this
tachyarrhythmia demonstrate a possible mechanism,
similar to parasystole, for the initiation of various
atrial tachyarrhythmias. The asynchronous depolarization of multiple pacemakers could be an underlying mechanism for several arrhythmias, such as premature atrial contractions, atrial parasystoles, and
multifocal atrial tachycardia, and could be the triggering mechanism for atrial flutter and atrial fibrillation. Two or more pacemakers depolarizing asynchronously and protected from resetting by entrance
block could generate a tachyarrhythmia even if none
of the pacemakers were tachycardic.
References
1. Lewis T: The Mechanism and Graphic Registration of the Heart
Beat, ed 3. London, Shaw & Sons, Ltd, 1925
2. Loomis TA, Krop S: Auricular fibrillation induced and maintained in animals by acetylcholine or vagal stimulation. Circ
Res 1955;3:390-396
3. James TN, Nadeau RA: Sinus bradycardia during injections
directly to the sinus node artery. Am J Physiol 1963;204:9-18
4. Allessie MA, Lammers WJEP, Bonke AM, Hollen J: Intraatrial reentry as a mechanism for atrial flutter induced by
acetylcholine and rapid pacing in the dog. Circulation 1984;70:
123-135
5. Nadeau RA, Roberge FA, Billette J: Role of the sinus node in
the mechanism of cholinergic atrial fibrillation. Circ Res 1970;
27:129-138
6. Boineau JP, Schuessler RB, Hackel DB, Miller CB, Brockus
CW, Wylds AC: Widespread distribution and rate differentiation of the atrial pacemaker complex. Am J Physiol 1980;239:
H406-H415
7. Randall WC, Talano J, Kaye MP, Euler D, Jones S, Brynjolfsson G: Cardiac pacemakers in absence of the SA node:
Responses to exercise and autonomic blockade. Am J Physiol
1978;234:H465-H470
8. Sealy WC, Seaber AV: Surgical isolation of the atrial septum
from the atria: Identification of an atrial septal pacemaker.
J Thorac Cardiovasc Surg 1980;80:742-749
9. Boineau JP, Schuessler RB, Roeske WR, Autry LJ, Miller CB,
Wylds AC: Quantitative relation between sites of atrial
impulse origin and cycle length. Am J Physiol 1983;245:
H781-H789
10. Hoffmnan BF, Cranefield PF: Electrophysiology of the Heart. Mt
Kisco, NY, Futura Publishing Co, Inc, 1960, pp 42-74
11. Schuessler RB, Bromberg BI, Boineau JP: Effect of neurotransmitters on the activation sequence of the isolated atrium.
Am J Physiol 1990;258:H1632-H1641
12. Bonke FIM, Allessie MA, Kirchoff CJHJ, Roos AGJM: Investigation of the conduction properties of the sinus node, in
Zipes DP, Jalife J (eds): Cardiac Electrophysiology andArrhythmias. Orlando, Fla, Grune & Stratton, Inc, 1985, pp 73-79
13. Rozanski GJ, Lipsius SL: Electrophysiology of functional
subsidiary pacemakers in canine right atrium. Am J Physiol
1985;249:H594-H603
14. Rosenshtraukh LV, Zaitsev AV, Fast VG, Pertsov AM, Krinsky VI: Vagally induced block and delayed conduction as a
mechanism for circus movement tachycardia in frog atria. Circ
Res 1989;64:213-226
15. Chiba S: Selective stimulation of intracardiac pre-ganglionic
vagal fibres of the dog atrium. Clin Exp Phannacol Physiol
1978;5:465-469
16. Sano T, Suzuki F, Sato S: Sinus node impulses and atrial
fibrillation. Circ Res 1967;21:507-513
17. Watanabe H, Perry JB, Page P, Savard P, Nadeau R: Vagal
effects of sinoatrial and atrial conduction studied with epicardial mapping in dogs: The influence of pacemaker shifts on the
measurement of sinoatrial conduction time. Can J Physiol
Pharmacol 1985;63:113-121
18. Cranefield PF, Aronson RS: Cardiac Arrhythmias: The Role of
Triggered Activity and Other Mechanisms. Mt Kisco, NY, Futura
Publishing Co, Inc, 1988, pp 125-160
Schuessler et al Spontaneous Cholinergic Tachyarrhythmia
19. Zaitsev AV, Rosenshtraukh LV, Fast VG, Bobrovskii RV,
Pertsov AM: Study of spontaneous acetylcholine-dependent
tachyarrhythmias using isolated specimens of the right canine
atrium by bilateral mapping of the spread of excitation.
Kardiologiia 1989;29:80-85
20. Boineau JP, Canavan TE, Schuessler RB, Cain ME, Corr PB,
Cox JL: Demonstration of a widely distributed atrial pacemaker
complex in the human heart. Circulation 1988;77:1221-1237
21. Bonke FIM, Kirchoff CJHJ, Allessie MA: Cellular function of
the sinus node during atrial fibrillation, in Attuel P, Coumel P,
Janse MJ (eds): The Atrium in Health and Disease. Mt Kisco,
NY, Futura Publishing Co, Inc, 1989, pp 65-78
1087
22. Coumel P: Clinical approach to paroxysmal atrial fibrillation.
Clin Cardiol 1990;13:209-212
23. Killip T, Gault JH: Mode of onset of atrial fibrillation in man.
Am Heart J 1965;70:172-179
24. Bennett MA, Pentecost BL: The pattern of onset and spontaneous cessation of atrial fibrillation in man. Circulation 1970;
41:981-988
KEY WORDS * acetylcholine * atrial fibrillation * pacemakers
* atrial tachyarrhythmias
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
Spontaneous tachyarrhythmias after cholinergic suppression in the isolated perfused
canine right atrium.
R B Schuessler, L V Rosenshtraukh, J P Boineau, B I Bromberg and J L Cox
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
Circ Res. 1991;69:1075-1087
doi: 10.1161/01.RES.69.4.1075
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1991 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/69/4/1075
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information
about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation Research is online at:
http://circres.ahajournals.org//subscriptions/