Vol. 274, Issue 3, H817-H828, March 1998
Modulation of quinidine-induced arrhythmias by temperature in
perfused rabbit heart
Joseph F.
Spear and
E. Neil
Moore
Department of Animal Biology, School of Veterinary Medicine,
University of Pennsylvania, Philadelphia, Pennsylvania 19104
 |
ABSTRACT |
We used low temperature to slow ion channel
kinetics and studied the electrophysiological effects of quinidine at
different pacing rates in isolated rabbit hearts. Fifteen epicardial
electrograms together with an endocardial monophasic action potential
were recorded. Epicardial activation and local recovery times were measured. Arrhythmias together with the characteristics of their mode
of induction and rate were analyzed by epicardial activation sequence
mapping. In the presence of quinidine, arrhythmias consistent with both
triggered activity and reentry were observed. At baseline, triggered
activity was not inducible, even though at 25°C the recovery time
was greater than that in the presence of quinidine at 36°C. Also,
with quinidine, the incidence of triggered activity decreased at 30 and
25°C. Therefore prolongation of the recovery time per se does not
cause triggered activity. Quinidine's use-dependent effects on
conduction and reverse use-dependent effects on recovery time were
amplified by low temperatures. These findings can be understood in
terms of the known temperature sensitivities of the kinetics of the
membrane ion channels responsible for activation and recovery. The
results demonstrate that temperature can be used as a tool to elucidate
mechanisms of drug action.
early afterdepolarizations; myocardial refractoriness; ventricular
tachycardia; use dependence; reverse use dependence
| |
INTRODUCTION |
RECENT CLINICAL TRIALS, such as the National Institutes
of Health-funded "Clinical Arrhythmia Suppression Trial," which
brought the efficacy of class I antiarrhythmic drug treatment into
question, have shifted attention to class III antiarrhythmic agents
(12). Their antiarrhythmic effect relies on prolonging myocardial
recovery by blocking K+ currents
associated with repolarization (6, 7, 10). An unwanted side effect is
that they may produce early afterdepolarizations (EADs), which can lead
to triggered beats. These triggered beats have been implicated in the
malignant arrhythmia torsade de pointes (14). Recent findings indicate
that the characteristic "twisting of points" seen in the body
surface electrocardiogram during the tachycardia is due to both
multifocal epicardial breakthroughs of triggered beats and also
triggered beats precipitating reentry (4, 13). Quinidine seems to be
particularly potent in causing torsade de pointes (11, 29). The
mechanism of the EADs is not well understood (19, 22). EADs are
bradycardia dependent. They appear after long coupling intervals that
prolong myocardial recovery. However, it is not clear whether they are
an obligatory side effect of prolonging myocardial recovery. Both the
endocardial Purkinje system and "M cells" of the midmyocardium
have been implicated as sites for the origin of EADs (1, 11). However,
recent studies suggest that, in the intact heart, they probably arise from the endocardium (3, 13). Two types of EADs are seen, those
occurring early during the plateau of the action potential and those
occurring later during repolarization (11). EADs do not always produce
triggered beats. Theoretically, whether a focal EAD can produce a
conducted response is determined by the electrical "load"
presented by the contiguous tissue and the tissue's excitability, and
there are potentially many foci competing for control. By their nature,
EADs are tied to the myocardium's repolarization phase, and this
raises the possibility that refractoriness of tissue adjacent to their
site of origin may be important in modulating whether an EAD is
manifest as a triggered ectopic beat. Also the conditions under which
triggered beats precipitate reentry are only beginning to be understood
(4, 13). We chose to study quinidine because of its dual
electrophysiological effect. Quinidine has both class I- and III-like
actions in that it slows conduction velocity as well as prolongs
recovery (30). Thus its potential for malignant proarrhythmia may
reside in this combination of actions.
It is well known that lowering myocardial temperature slows the
kinetics of the membrane ion channels responsible for activation and
recovery, and their relative temperature sensitivities vary (9, 24, 25,
26). The current study was designed to take advantage of this and
investigate how changes in myocardial conduction and refractoriness
induced by temperature alterations modulates quinidine's
arrhythmogenic effects with regard to both myocardial triggered
activity and reentry. We found that alterations in temperature can be a
useful tool in elucidating mechanisms of drug action.
| |
METHODS |
Langendorff-perfused rabbit heart.
All experiments conformed to the National Institutes of Health
"Guide for the Care and Use of Laboratory Animals" (DHHS
publication no. NIH 86-23 revised 1985) and were approved by the
Institutional Animal Care and Use Committee of the University of
Pennsylvania.
Twelve adult New Zealand White rabbits of either sex (2.5-3.5 kg)
were sedated with a combination of ketamine hydrochloride (40 mg/kg)
and xylazine (4.4 mg/kg) administered intramuscularly. Sedated rabbits
were heparinized (500 U iv) and killed with pentobarbital sodium (25 mg/kg iv). The heart was removed via a thoracotomy and rinsed in
oxygenated Tyrode solution. The aorta was cannulated, and the heart was
hung on a Langendorff perfusion apparatus in which the coronary
arteries were perfused with oxygenated, temperature-controlled Tyrode
solution at a constant pressure of 80 mmHg. Constant pressure was
provided by a column of perfusate of fixed height above the aortic
cannula and was surrounded by a water jacket. This produced a coronary
flow of ~30 ml/min, which did not vary by >10% during the time of
the experiment. A single controller was used to control and vary the
temperature of the water jacket. The millimolar composition of the
Tyrode solution was 137.0 NaCl, 24.0 NaHCO3, 5.5 dextrose, 2.7 KCl, 2.7 CaCl2, 0.9 NaH2PO4,
and 0.5 MgCl2. Myocardial
temperature was monitored with a thermistor probe placed within the
right or left ventricle. Experiments commenced after an equilibration period of 1 h.
Quinidine was introduced directly into the perfusate from a stock
solution to attain concentrations of 2 or 4 mg/l (2.5 or 5.0 mM). The
concentration of the quinidine stock solution was 40 mg/ml and was made
fresh weekly by dissolving quinidine sulfate (Sigma Chemical) in 50%
ethanol and water.
Electrophysiological methods.
To slow the heart rate for pacing, atrioventricular block was induced
by ablating the bundle of His using a portable battery-powered cautery.
A flexible silicone rubber plaque of 15 bipolar (2.5-mm separation),
0.5-mm-diameter silver electrodes arranged in a three by five array
with a 7.0-mm interelectrode spacing was wrapped around the right and
left ventricles. The left anterior descending coronary artery was used
as a landmark to facilitate orienting the electrodes similarly for each
heart. The silicone rubber electrode array had an added advantage in
that it acted as a temperature insulator and channeled the perfusate
exiting the right atrium to flow between it and the ventricular
epicardium. A monophasic action potential recording catheter (15, 28)
was positioned in either the right or left ventricular endocardium at
its apex. The bipolar electrograms were simultaneously recorded
together with the monophasic action potential.
The heart was paced by constant-current, 2.0-ms-duration pulses
delivered at twice diastolic threshold intensity through a pair of
close bipolar electrodes embedded in the plaque and positioned over the
epicardium close to the interventricular septum.
Electrograms and the monophasic action potential were amplified using
custom-designed amplifiers with a bandwidth of 0.1-1,000 Hz. The
amplified electrograms were recorded on a strip-chart recorder at a
paper speed of 100 mm/s. Local epicardial activation time was
determined by measuring the time of the occurrence of the intrinsic
deflection of the electrogram relative to the pacing artifact using a
manual digitizer tablet (GTCO, Rockville, MD) interfaced with a
computer (Hewlett-Packard, Sunnyvale, CA). The resolution of the system
was 0.1 ms. The intrinsic deflection of an electrogram was taken as the
time of the most rapid deflection. The recovery time for each
electrogram was taken as the time from the stimulus artifact to the
most rapidly changing late part of the T wave. It has been shown that
this provides a measure of local action potential repolarization time
and can be used to track changes induced by interventions (17). The
activation-recovery interval was defined as the difference between an
electrogram's recovery time and activation time. This provides an
index of local recovery independent of activation time. The SD of the
mean activation-recovery interval of all 15 electrograms was used as an
index of the intrinsic heterogeneity of recovery time in the heart.
The Q10 for each parameter was
calculated to define its temperature sensitivity. We defined it as the
ratio of the magnitude of the parameter before and after a 10°C
change in temperature.
Experimental protocols.
In a previous study, we found that the electrophysiological
characteristics of our preparation remained steady over a period of at
least 3 h (32). In the present study, our protocols lasted from 1.75 to
3 h (a mean of 2.4 h). After the equilibration period, baseline
measurements were made. Electrograms and monophasic action potentials
were recorded during pacing at basic cycle lengths of 600, 700, 1,000, and 2,000 ms and at perfusion temperatures of 25, 30, and 36°C.
After baseline measurements, the hearts were perfused with Tyrode
solution at 36°C containing quinidine sulfate at a concentration of
2 mg/l. Within 5-15 min, spontaneous, closely coupled beats appeared after both paced and spontaneous beats occurring at long basic
cycle lengths. In 3 of the 12 hearts, no spontaneity appeared at 2 mg/l; therefore the concentration was increased to 4 mg/l. In these
hearts, the effects were not distinguishable from those given 2 mg/l;
so both groups were analyzed together. As soon as close-coupled beats
appeared, the heart was paced at a basic cycle length of 
2,000 ms,
and ~40-60 sequential paced beats were recorded together with
any associated spontaneous activity. This was followed by pacing at
basic cycle lengths of 600, 700, 1,000, and 2,000 ms to evaluate
activation-recovery intervals in the presence of quinidine. We found
that if we paced the heart at the short basic cycle lengths initially,
it facilitated being able to obtain data at the long cycle lengths with
less interference by closely coupled ectopic beats, which confound the
measurement of recovery time. Subsequently, the temperature of the
perfusate was reduced to 30 and 25°C, and the pacing protocols were
repeated at those temperatures. The temperature was then returned to
36°C to verify that ectopy was similar to that during the initial
period at this temperature. It took ~10-15 min for the
temperature to stabilize at the new values. The range of the pacing
basic cycle lengths that we used were chosen to allow us to pace at
each temperature without interference from spontaneous automatic beats
at the long cycle lengths or being limited by refractoriness with the
short cycle lengths. However, there were still cases in which we were
unable to obtain data because of our inability to reliably capture the
ventricles due to interference by spontaneous automaticity or
arrhythmias. We found that the effects of quinidine did not completely
resolve even after 1 h when we changed the perfusate back to the normal Tyrode solution. Therefore measurements after washout of quinidine are
not reported.
Data analysis.
During baseline and after quinidine at each temperature and pacing
basic cycle length, the local activation times as indicated by the
intrinsic deflections and times of recovery as indicated by the local T
waves were digitized for each electrogram and tabulated.
The epicardial electrograms were also used to generate activation
sequence maps, which together with additional criteria allowed us to
characterize the spontaneous rhythms. We characterized type I
arrhythmias by the following criteria. The activation sequence maps
showed a focal origin with a radial spread of activity from that
region. The extra beat was closely coupled to the recovery time of the
previous beat. (This is unlike that of spontaneous automatic
ventricular beats, which, though focal in origin, occur with long
coupling intervals usually >2,000 ms.) The extra beats followed beats
having long coupling intervals (2,000 ms). These characteristics are
consistent with arrhythmias being due to triggered activity (4, 11,
13).
We characterized type II arrhythmias using the following criteria. The
activation sequence maps showed a circuitous activation sequence. The
circuitous conducting beats followed after a locally blocked beat. The
extra beats followed beats with short coupling intervals (
400 ms).
These characteristics are consistent with the arrhythmias being due to
reentry (4, 13).
To quantify the incidence and coupling intervals of type I and II beats
occurring in the presence of quinidine, we analyzed the 40-60
sequential paced beats in each heart which were delivered at cycle
lengths of 2,000 ms. Paced beats showing no close-coupled ectopy as
well as the number and coupling intervals of close coupled ectopic
beats after a paced beat (type I) were noted. If type II activity
occurred, its mean cycle length and the coupling interval of the
precipitating beat were noted.
Results are presented as means ± SD. Differences in measured
variables between baseline and quinidine were compared by a two-tailed Student's t-test for paired data.
Differences at the three temperatures were compared using an analysis
of variance (ANOVA). P < 0.05 was
considered statistically significant. Data involving arrhythmia characteristics were obtained on 12 hearts. However, due to problems with pacing the preparation, we obtained activation and recovery data
in only 10 of the 12 hearts. In the tables containing the activation
and recovery data, the number of hearts reported varied from 5 to 10. This was due to our inability at some temperatures and heart rates to
adequately pace the ventricles, as mentioned previously.
| |
RESULTS |
Effect of quinidine on myocardial activation and recovery.
All activation and recovery data were obtained during pacing from the
same relative left ventricular epicardial site. Figure 1 displays selected electrograms and
activation sequence maps during pacing at a basic cycle length of 2,000 ms for two conditions representing the extremes of our protocol. In
Fig. 1A, the data were obtained at a
perfusion temperature of 36°C at baseline before the addition of
quinidine. In Fig. 1B, the data were
obtained at 25°C in the presence of quinidine. The vertical arrows
above the electrograms show the points used to determine recovery
times. Notice that even with the extreme prolongation of activation and recovery in Fig. 1B, points of
measurement with configurations comparable to Fig.
1A could be obtained. Also, notice
in the maps below that although conduction velocity was significantly
slowed in Fig. 1B, the activation
sequence was nearly identical to that in Fig.
1A. This behavior was similar in all
of our hearts and allows us to use changes in mean activation time as a
direct index of changes in mean conduction velocity.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 1.
Examples of epicardial electrograms
(top) and activation sequence maps
(bottom) obtained at pacing basic
cycle lengths of 2,000 ms at 36°C at baseline
(A) and at 25°C in presence of
2.0 mg/l quinidine (B). Brief spikes
at left of each electrogram's intrinsic deflection are stimulus
artifacts. Vertical arrow above each electrogram indicates point where
recovery time was measured. In activation sequence maps
(bottom), borders of posterior left
ventricle (Post LV), anterior left ventricle (Ant LV), and right
ventricle (RV) are delineated by dotted lines. LAD, left anterior
descending coronary artery. Numbers are activation times (in ms)
relative to time of stimulus artifact.
|
|
In Fig. 2 are presented the mean activation
times (graphs at top) and
activation-recovery intervals (graphs at
bottom) at four paced heart rates
and at perfusion temperatures of 36, 30, and 25°C (Fig. 2,
A-C, respectively). At the
relatively low concentration used, quinidine had a small effect on mean
activation time. However, "use dependence" (20) is apparent,
since the effect was greater at the faster heart rates. Also, both the
prolongation of mean activation time and the use dependence was
augmented at the lower temperature.
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 2.
Graphs showing effect of heart rate and temperature on mean activation
times (top) and mean
activation-recovery intervals
(bottom) at baseline ( ) and in
presence of quinidine ( ). Data in A
were obtained at a temperature of 36°C, in
B at 30°C, and in
C at 25°C. Vertical bars, SD.
* Significant difference by paired
t-test between baseline data and those
obtained in presence of quinidine.
|
|
Quinidine caused a significant prolongation of the activation-recovery
interval at all heart rates and temperatures. However, the degree of
prolongation was enhanced at the lower temperatures. The decrease in
the magnitude of quinidine's effect on recovery at the faster heart
rates that we observed especially prominently at 25°C has been
referred to as "reverse use dependence" (10, 21). Quinidine also
increased the SD of the activation-recovery interval and therefore the
heterogeneity of recovery (Table 1). The
effect of heart rate and temperature on heterogeneity was inconsistent.
However, it tended to be increased at slower heart rates and at lower
temperatures.
The Q10 of an effect characterizes
its temperature sensitivity. Table 2 shows
the Q10 for each of the above
parameters at the four heart rates. The
Q10 for the activation time tended
not to change with increasing heart rate, whereas the
Q10 for the activation-recovery
interval and its SD went down. The
Q10 in the presence of quinidine
was not significantly different from baseline in most cases.
Table 3 presents the effects of quinidine
on the left ventricular endocardial monophasic action potential
duration measured from the rapid upstroke to 90% repolarization. The
results parallel those of the effect of temperature and quinidine on
the activation recovery interval of the epicardial electrograms (Fig.
2B). However, the endocardial action
potential duration was prolonged to a greater degree at 25°C than
the epicardial activation-recovery interval (P = 0.03). Also, the effect of
quinidine on the endocardial action potential duration was greater at
all temperatures (P 0.004).
Characteristics of ectopic activity induced by quinidine.
All hearts exhibited quinidine-induced arrhythmias, and their
occurrence was preceded by the appearance of early afterdepolarizations in the endocardial monophasic action potential (MAP) recording. The
arrhythmias began 5-15 min after addition of quinidine to the
perfusate. Type I activity (up to 12 consecutive beats) was observed in
each heart, and type II activity (up to 7 consecutive beats) induced by
type I activity was observed in 10 of the 12 hearts. Figure
3A shows
an example of type I activity leading to type II activity. There were
11 beats in the tachycardia, and the cycle length of the tachycardia
progressively shortened before termination. In Fig.
3B are epicardial maps for
beats 1,
2, 6, and 7. The map of
beat 1 is referenced to the upstroke
of the left ventricular monophasic action potential.
Beat 1 was a spontaneous beat that
occurred after a pause of 3 s and had an early epicardial breakthrough
near the interventricular septum. It activated the entire anterior left
ventricle nearly simultaneously. Beats with these activation
characteristics occurred very consistently throughout the experiment
and were not always followed by extra beats. On the basis of this, we
interpret beat 1 as due to an
automatic focus originating from the left ventricular endocardium. The
map of beat 2 showed an early
breakthrough on the right ventricular base (0-ms reference) to activate
right and left ventricles. Beats 3-5 showed similar characteristics but occurred at
progressively shorter cycle lengths. Beat
6 (referenced to earliest epicardial breakthrough)
showed a line of block at sites 11 and
15 and precipitated a type II
arrhythmia, which persisted from beats
7 through 11. We
interpret this rhythm sequence as a spontaneous left ventricular automatic beat after a pause causing trigger beats originating in the
right ventricle. The shortening coupling intervals of the triggered
beats eventually compromised conduction and caused local unidirectional
block precipitating reentrant tachycardia.
View larger version (47K):
[in this window]
[in a new window]
|
Fig. 3.
An example of type I activity leading to type II activity. In
A are records obtained during a
spontaneous tachycardia of 11 beats. Each beat is numbered above, and
their coupling intervals (in ms) are indicated. At
left, epicardial electrograms are
numbered from 1 to 15. At bottom is a
left ventricular endocardial monophasic action potential (MAP)
recording. In B are activation
sequence maps for beats 1,
2, 6,
and 7. Recording sites corresponding
to electrograms in A are
1-3 from
bottom to
top on Post LV and
4-15 from bottom
to top of columns,
right to left, on Ant LV and RV.
|
|
The role of refractoriness in determining local block of type I beats
is demonstrated in Fig. 4, which shows that
it is the local time of recovery that determines the region of block
for the beat. This preparation exhibited a stable and reproducible automatic focus that was occasionally accompanied by closely coupled, type I beats. This allowed us to define the recovery characteristics of
the heart after the automatic beat and how this influenced the closely
coupled beat. Figure 4A shows the
activation sequence of the spontaneous automatic beat (referenced to
left ventricular endocardium activation). Figure
4B shows this beat's recovery times
at each site. The shaded area indicates the sites with the longest
recovery times (445-461 ms). Figure
4C (referenced to left ventricular
endocardium activation of automatic beat) shows the activation sequence
of a relatively late coupled type I beat (earliest activation at 487 ms). In Fig. 4D is shown a similar beat but with an earliest activation of 322 ms. This beat blocked in
the area exhibiting the longest recovery interval.
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 4.
Role of local refractoriness in precipitating block of a type I beat.
A and
B: maps demonstrating activation
sequence for a spontaneous automatic beat and its corresponding local
recovery times, respectively. C and
D: activation sequence maps for a type
I beat that conducted throughout heart and one that blocked regionally,
respectively. Shading in maps in B and
D indicates sites with longest
recovery times.
|
|
Effect of temperature on occurrence and coupling intervals of
ectopic activity.
Table 4 presents the number of type I beats
that occurred after beats with long cycle lengths and number of type II
beats that were induced by type I activity during our observation
periods at each temperature. As the temperature was decreased, both the incidence and number of type I beats decreased. At 25°C, only 69 of
638 beats with long cycle lengths produced type I beats. Of the 779 episodes of type I activity that were logged at 36 and 30°C, 78 (10%) precipitated type II activity. Type II activity occurred after
from 1 to 10 (mean of 1.9) type I beats. Type II activity was not
observed at 25°C. This was probably due to the lack of spontaneous
close coupled type I beats, since in several hearts we were able to
induce reentry under these conditions by paced premature ventricular
beats.
In Table 5, the mean cycle lengths of the
type I and II beats are shown. At the lower temperatures, the mean
cycle lengths of the type I beats were significantly prolonged
(P < 0.001 by ANOVA) from 366.0 ± 61.2 ms at 36°C to 649.2 ± 126.2 ms at 30°C and
1,225.8 ± 344.9 ms at 25°C. This was not observed to occur with
the cycle lengths of the type II beats. Not only was the cycle length
of the type II activity at 36°C significantly shorter (P < 0.001) than for the type I
beats (267.1 ± 39.2 ms), but also it did not significantly prolong
at 30°C (287.6 ± 84.5 ms, P = 0.51 by ANOVA).
Modulation of cycle length of type I beats by myocardial recovery.
Figure 5,
A and
B, presents the relationship between
the mean activation-recovery interval during paced rhythms and the mean cycle length of the type I and II beats, respectively, at three temperatures. Because type I beats occurred after beats with long coupling intervals, the mean activation-recovery intervals are those
obtained during pacing at a cycle length of 2,000 ms (Fig. 5A), and because type II activity
was induced by short coupling intervals, the mean activation intervals
in Fig. 5B are those obtained during
pacing at a cycle length of 600 ms. Although the mean cycle length of
the type I activity tracked the prolongation of the activation-recovery
interval as temperature was lowered (Fig.
5A), this did not occur with the
type II activity (Fig. 5B).
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 5.
Graphs relating mean cycle lengths of type I and II activity
(A and
B, respectively) to mean
activation-recovery intervals at temperatures of 36, 30, and 25°C.
For type I activity, mean activation recovery intervals were those
obtained at a basic cycle length of 2,000 ms. For type II activity,
they were those obtained at a basic cycle length of 600 ms. Type II
activity did not occur at 25°C.
|
|
Concealed type I activity.
Not only does myocardial recovery determine the coupling interval of
manifest type I beats, but also it may contribute to the relative
disappearance of this activity at 25°C. Type I beats may block
locally as observed in Figs. 3 and 4, and this may lead to type II
activity. We observed that, at low temperature when recovery was
prolonged, there were occasions on which type I beats were concealed
within the endocardium, and this may play a role in the decreased
incidence of manifest type I beats at lower temperatures. Figure
6 presents records obtained in a
preparation during perfusion with 4 mg/l quinidine at a temperature of
30°C. In Fig. 6A, an extra beat is
shown completely concealed within the left ventricular endocardium. It
became manifest (Fig. 6B) when the
coupling interval of the extra beat was delayed by 18 ms.
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 6.
Records obtained during perfusion with 4 mg/l quinidine at 30°C,
demonstrating a possible concealed focal triggered beat. In
A are electrograms recorded during a
spontaneous beat that occurred after a pause of at least 8 s. Trace at
bottom is a left ventricular
endocardial MAP recorded near apex. Notice in MAP recording that a
rapid intrinsic deflection followed an early afterdepolarization (open
arrow). Electrograms did not show corresponding rapid deflections but
only simultaneous slow, low-amplitude waves (indicated by question
mark). In B, records were obtained
during a stimulated (S) beat after a pause of 4 s. In this case rapid
intrinsic deflection after early afterdepolarization (open arrow) was
delayed by 18 ms compared with that seen in
A and was apparently able to produce a
conducted beat (solid arrow), which was followed by a second triggered
beat.
|
|
Another example in Fig. 7 shows a sequence
in which the extra beat originated at a distance from the MAP recording
site. At the shorter coupling interval (Fig.
7A), not only was the activity largely excluded from the epicardium, but also it did not engage the
endocardial region of the MAP recording presumably due to its
refractoriness. However, the MAP repolarization time was greatly prolonged probably due to electrotonic current flowing from adjacent active regions. At the longer coupling interval (Fig.
7B), rapid deflections
appeared on the MAP recording (open arrow), and epicardial activation also became apparent (closed arrow).
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 7.
Records demonstrating a type I beat that did not activate region of a
locally recorded left ventricular endocardial MAP. Data were obtained
at 27.2°C in presence of 2.0 mg/l quinidine. In
A, extra beat had a coupling interval
of ~760 ms (indicated by question mark) and did not activate region
of MAP recording but did prolong duration of MAP. In
B, extra beat activated epicardium at
a coupling interval of 1,300 ms (solid arrow), and an intrinsic
deflection appeared during plateau of MAP (open arrow).
|
|
| |
DISCUSSION |
Our studies demonstrate that temperature-induced slowing of ion channel
kinetics can be used as a tool to elucidate mechanisms of drug action.
Our major findings are as follows. Prolongation of the recovery time of
the myocardium per se does not cause type I triggered activity. The
coupling intervals of manifest triggered beats caused by quinidine are
determined by the myocardium's refractoriness, and prolonging the
recovery time of the myocardium decreases the incidence of triggered
activity. Quinidine's use-dependent effects on conduction and reverse
use-dependent effects on recovery are amplified at low temperatures.
This temperature dependence of quinidine's effect on the myocardium's
activation-recovery interval is compatible with the idea that the
delayed rectifier K+ channel
contributes importantly to prolongation of action potential duration at
low temperature and that quinidine's reverse use dependence is due to
a reduced diastolic interval at faster heart rates.
Temperature modification as a tool to elucidate mechanisms of drug
action.
There are several possible mechanisms that can contribute to action
potential prolongation at low temperature. Studies in guinea pig
myocytes indicate that the major factor is a decrease in the delayed
rectifier K+ current
(IK, primarily
fast component), which exhibits a temperature coefficient
(Q10) of 4.4 (26). Delayed
inactivation of the L-type Ca2+
current and inward rectifier K+
current also contribute with Q10
of 2.3 and 1.5, respectively (9, 26). Both activation and inactivation
time constants of the fast Na+
channel have a Q10 of 3 (24). This
temperature sensitivity contributes to conduction velocity slowing at
low temperatures. However, slowing of the inactivation of the
Na+ current appears not to be an
important factor in the prolongation of action potential duration at
low temperature (25). A time-independent current that is temperature
sensitive has been described. Although it is not due to an electrogenic
Na+-K+
pump, its exact nature is unknown (26). The transient outward current
(Ito)
contributes to repolarization in rabbit myocardium (21, 18), especially
at rates slower than 2 Hz (31). However, its temperature sensitivity
has not been defined.
Because of these differences in the temperature sensitivity of various
channels and the fact that slowing their kinetics may have either
opposing or synergistic effects dependent on how they contribute to a
given electrophysiological phenomenon, we can use
temperature to dissect out mechanisms contributing to drug action.
Characteristics of quinidine-induced arrhythmias.
We found that the spontaneous arrhythmias in the isolated rabbit heart
were consistent with both triggered activity and reentry induced by
triggered activity. Similar findings have been reported in the isolated
rabbit heart (4) and in vivo canine heart (13). At 36°C (Table 5),
reentrant beats (type II) have shorter coupling intervals than
triggered beats (type I). This reflects their contrasting mechanisms.
Triggered activity follows beats having long coupling intervals and
therefore long recovery times. Reentry is induced by premature
activations with short coupling intervals and therefore short recovery
times. Also, the short coupling intervals are maintained by the rapidly
cycling reentrant wave front.
At reduced temperatures, the incidence of both type I and II activity
was attenuated (Table 4). As can be seen in Table 1, at 25°C the SD
of the activation-recovery interval (heterogeneity) was increased by
quinidine. This would favor reentry occurring. If the triggered beats
have sufficiently short coupling intervals to block locally, then
reentry might be induced (Figs. 3 and 4). The disappearance of
reentrant activity at 25°C is probably related to the reduction in
the occurrence of the precipitating events (triggered beats).
The mechanism for the decreased occurrence of type I activity at low
temperature is probably related primarily to temperature-induced changes in membrane ion channels as discussed below. However, extreme
prolongation of refractoriness may also be involved. Figures 6 and 7
show examples of triggered beats that did not break through to activate
the epicardium. It is possible that the actual frequency of occurrence
of type I beats was underestimated in our observations because they
were concealed near their site of origin.
Role of prolongation of recovery in quinidine-induced arrhythmias.
Table 5 and Fig. 5 show that the coupling intervals of type I beats are
modulated by the time of myocardial recovery. This is expected, since
EADs are associated with the repolarization phase of the action
potential, and for a triggered beat to be manifest, it must be able to
conduct away from its site of origin in recovered tissue.
The facts that prolongation of the activation-recovery interval by low
temperature does not of itself induce early afterdepolarizations and
that quinidine-induced triggered activity is attenuated by low
temperature indicate prolongation of recovery per se is not the direct
causal factor in the triggered arrhythmias. There is evidence that some
EADs are due to recovery and reactivation of L-type
Ca2+ channels during the plateau
of prolonged action potentials and that the plateau must be in a
voltage range from which recovery from inactivation and reactivation
can occur (22). One mechanism for the antagonism of EAD occurrence
during prolongation of the action potential by low temperature may be
that it places the plateau in a nonoptimal voltage range and/or
changes the voltage dependence of the
Ca2+ window current responsible
for the EAD. These possibilities can be tested experimentally and may
have implications for designing drugs that prolong recovery but do not
induce EADs.
Mechanism of quinidine's effect on myocardial activation and
recovery.
Figure 2, top, demonstrates
quinidine's effect on conduction at three different temperatures and
over a range of heart rates. Quinidine slows conduction by binding to
the active state and blocking the fast
Na+ channel (20). At 36°C
(Fig. 2A) at the doses that we used, there was a relatively small effect of quinidine on the mean activation time, and there was no evidence of use dependence (20). At 30 and
25°C (Fig. 2, B and
C, respectively), the baseline
conduction times were significantly prolonged due to low
temperature-induced slowing of the
Na+ channel kinetics. The effect
of quinidine under the conditions of slowed
Na+ channel kinetics was enhanced,
possibly due to increased drug binding because of the increased time
the channel spent in the active state. Use dependence became prominent
under these conditions. Thus low temperature amplified a subtle drug
effect.
Quinidine also blocks the delayed rectifier
K+ channel (6, 8, 10, 23, 34).
This prolongs the action potential duration and has also been
implicated in producing early afterdepolarizations and triggered
activity (11, 29). Figure 2, bottom,
demonstrates quinidine's effect on the activation-recovery interval.
At 36°C (Fig. 2A), quinidine
prolonged the activation-recovery interval at all heart rates. Low
temperature (Fig. 2, B and
C) prolonged the baseline
activation-recovery interval, and as for the mean activation time, it
amplified the effect of quinidine on the activation-recovery interval
and potentiated the reverse use-dependent effect associated with faster
heart rates. This was also observed for the endocardial MAP (Table 3).
Interestingly, triggered beats were prominent with quinidine at
36°C but never occurred under baseline conditions and were
practically absent in the presence of quinidine at 25°C (Table 4).
The mechanism of the reverse use dependence of agents that block the
delayed rectifier currents is controversial (5, 7, 18, 21, 23). The
activation-recovery interval normally shortens at faster heart rates,
and this is presumably due to incomplete recovery of plateau currents
(16, 31). This also occurs in the presence of drug-induced prolonged
recovery due to delayed rectifier
K+ channel block. Recent data
indicate that quinidine binds to the delayed rectifier
K+ channel in the resting state
(34). Because there is evidence that the degree of channel binding is
not attenuated at the faster heart rates (5, 7), it has been suggested
that the mechanism for reverse use dependence is the normal
interval-dependent incomplete recovery of plateau currents (7) or
involvement of the slowly activating component of the delayed rectifier
K+ channel (23). Also, in the
rabbit heart, Ito
plays a prominent role in determining action potential duration (18).
Data indicate that sematilide's reverse use-dependent effect may be
due to activation of the
Ca2+-sensitive component of
Ito by increased
intracellular Ca2+ at faster heart
rates (5). Other factors that may contribute to rate-dependent
reduction in action potential duration are extracellular accumulation
of K+ (27) and the electrogenic
Na+-K+
pump (33). The data of Fig. 2, bottom,
and Tables 2 and 3, which demonstrate the temperature dependence of
reverse use dependence, can shed some light on its mechanism.
It is the diastolic interval between the preceding recovery time and
the subsequent activation that determines that action potential's
duration. This is the time available for recovery of the plateau
currents (16, 31). This interval is greatly reduced for a given heart
rate change when the time of repolarization has been prolonged by low
temperature. The prolongation of repolarization is due primarily to a
slowing in the activation of the delayed rectifier channel, which is
relatively temperature sensitive with a
Q10 of 4.4 (26), and delay in the
inactivation of
Ito (18). In
addition, the recovery kinetics of the delayed rectifier channel must
also be slowed. Thus, at low temperature, the interval-dependent shortening was enhanced as demonstrated in Fig. 2 by the steeper relationship between the activation-recovery interval and heart rate
under baseline conditions at the lower temperatures. When repolarization was further prolonged by blocking the delayed rectifier channels, this resulted in a much greater relative reduction in the
diastolic interval at the lower temperatures. This would account for
the greater effect of quinidine on the activation-recovery interval for
a given heart rate at the lower temperatures (cf. Fig. 2,
A-C). However, this does not
account for quinidine's greater reverse use-dependent effect
(relatively steeper slope of relationship between activation-recovery
interval and heart rate) with lower temperatures.
Our data indicate that factors such as incomplete recovery of plateau
currents (7), accumulation of extracellular
K+ (27), electrogenic
Na+-K+
pumping (33), activation of
Ito (5), and the
slow component of
IK (23) probably
do not contribute importantly to quinidine's reverse use dependence.
Low temperature-induced slowing of the kinetics of any of these
mechanisms would tend not to change the slope of the relationship
between the activation-recovery interval and heart rate or would tend
to decrease the slope of the relationship.
Our data are compatible with quinidine's reverse use dependence being
due to a reduced time available for binding to the resting state of the
delayed rectifier channel at faster heart rates (34). The increase in
slope of the relationship between the activation-recovery interval and
heart rate at low temperature is explained in these terms. If quinidine
binds to the resting state of the channel, the relatively greater
reduction in diastolic interval for a given heart rate change that
occurs at the lower temperatures would allow even less time for binding
and account for the exaggeration of reverse use dependence at lower
temperatures. The decrease in the degree of channel block at faster
heart rates is also compatible with the apparent decrease in the
Q10 of quinidine's effect at faster heart rates (Table 2).
Limitations.
Even in the relatively small rabbit heart, the resolution of our
recordings limits the degree to which we can localize the electrophysiological phenomena under study. We utilized only one endocardial monophasic action potential recording together with the
epicardial electrograms; therefore the exact focal origin of the
triggered beats could not be identified. Also, we could not determine
whether locally concealed triggered activity (as exemplified by Figs. 6
and 7) is a common occurrence. Our 15-site electrode array on
a 2-cm-diameter rabbit heart provides the interelectrode spacing comparable to 240 sites on an 8-cm-diameter canine heart. Although the interelectrode spacing of our epicardial recordings allowed us to reliably distinguish type I (triggered) beats from type
II (reentrant) beats, it limited our ability to finely correlate ectopic activity with local refractoriness. We were limited to the more
general correlations as shown in Figs. 4 and 5.
| |
FOOTNOTES |
Address for reprint requests: J. F. Spear, Suite 201E, Dept. of
Animal Biology, School of Veterinary Medicine, 3800 Spruce St.,
Philadelphia, PA 19104-6046.
Received 30 May 1997; accepted in final form 14 November 1997.
| |
REFERENCES |
1.
Antzelevitch, C.,
and
S. Sicouri.
Clinical relevance of cardiac arrhythmias generated by afterdepolarizations. Role of M cells in the generation of U waves, triggered activity, and torsade de pointes.
J. Am. Coll. Cardiol.
23:
259-277,
1994[Abstract].
2.
Antzelevitch, C.,
S. Sicouri,
S. H. Litovsky,
A. Lukas,
S. C. Krishnan,
J. M. Di Diego,
G. A. Gintant,
and
D.-W. Liu.
Heterogeneity within the ventricular wall: electrophysiology and pharmacology of epicardial, endocardial and M cells.
Circ. Res.
69:
1427-1449,
1991[Free Full Text].
3.
Anyukhovsky, E. P.,
E. A. Sosunov,
and
M. R. Rosen.
Regional differences in electrophysiological properties of epicardium, midmyocardium, and endocardium
in vitro and in vivo correlations.
Circulation
94:
1981-1988,
1996[Abstract/Free Full Text].
4.
Asano, Y.,
J. M. Davidenko,
M. S. Baxter,
R. A. Gray,
and
J. Jalife.
Optical mapping of drug induced polymorphic arrhythmias and torsade de pointes in the isolated rabbit heart.
J. Am. Coll. Cardiol.
29:
831-842,
1997[Abstract].
5.
Beatch, G. N.,
D. R. Davis,
S. Laganiere,
and
B. A. Williams.
Rate-dependent effects of sematilide on ventricular monophasic action potential duration, and delayed rectifier K+ current in rabbits.
J. Cardiovasc. Pharmacol.
28:
618-630,
1996[Medline].
6.
Carmeliet, E.
Electrophysiologic and voltage clamp analysis of the effects of sotolol on isolated cardiac muscle and Purkinje fibers.
J. Pharmacol. Exp. Ther.
232:
817-825,
1985[Abstract/Free Full Text].
7.
Carmeliet, E.
Voltage and time-dependent block of the delayed rectifier K+ current in cardiac myocytes by dofetilide.
J. Pharmacol. Exp. Ther.
262:
809-817,
1992[Abstract/Free Full Text].
8.
Carmeliet, E.
Use-dependent block of the delayed K+ current in rabbit ventricular myocytes.
Cardiovasc. Drugs Ther.
7:
599-604,
1993.
9.
Cavalie, A.,
T. F. McDonald,
D. Pelzer,
and
W. Trautwein.
Temperature-induced transitory and steady-state changes in the calcium current of guinea-pig ventricular myocytes.
Pflügers Arch.
405:
294-296,
1985[Medline].
10.
Colatsky, T. J.,
W. Spinelli,
and
I. F. Moubarak.
Block of myocardial potassium channels by antiarrhythmic drugs: dependence on channel gating.
In: Ion Channels in the Cardiovascular System. Function and Dysfunction, edited by P. M. Spooner,
and A. M. Brown. Armonk, NY: Futura, 1994, p. 414-423.
11.
Davidenko, J. M.,
L. Cohen,
R. Goodrow,
and
C. Antzelevitch.
Quinidine-induced action potential prolongation, early afterdepolarizations, and triggered activity in canine Purkinje fibers. Effects of stimulation rate, potassium, and magnesium.
Circulation
79:
674-686,
1989[Abstract/Free Full Text].
12.
Echt, D. S.,
P. R. Liebson,
L. B. Mitchell,
R. W. Peters,
D. Obias-Manno,
A. H. Barker,
D. Arensberg,
A. Baker,
L. Friedman,
H. L. Greene,
M. L. Huther,
and
D. W. Richardson, CAST Investigators.
Mortality and morbidity in patients receiving encainide, flecainide, or placebo.
N. Engl. J. Med.
324:
781-788,
1991[Abstract].
13.
El-Sherif, N.,
E. B. Caref,
H. Yin,
and
M. Restivo.
The electrophysiological mechanism of ventricular arrhythmias in the long Q-T syndrome, tridimensional mapping of activation, and recovery patterns.
Circ. Res.
79:
474-492,
1996[Abstract/Free Full Text].
14.
El-Sherif, N.,
R. H. Zeiler,
W. Craelius,
W. B. Gough,
and
R. Henkin.
QTU prolongation, and polymorphic ventricular tachyarrhythmias due to bradycardia-dependent early afterdepolarizations. Afterdepolarizations and ventricular arrhythmias.
Circ. Res.
63:
286-305,
1988[Abstract/Free Full Text].
15.
Franz, M. R.
Long term recording of monophasic action potentials from human endocardium.
Am. J. Cardiol.
51:
1629-1634,
1983[Medline].
16.
Hauswirth, O.,
D. Noble,
and
R. W. Tsien.
The dependence of plateau currents in cardiac Purkinje fibers on the interval between action potentials.
J. Physiol. (Lond.)
222:
27-49,
1972[Abstract/Free Full Text].
17.
Haws, C. W.,
and
R. L. Lux.
Correlation between in vivo transmembrane action potential durations and activation-recovery intervals from electrograms. Effects of interventions that alter repolarization time.
Circulation
81:
281-288,
1990[Abstract/Free Full Text].
18.
Hiraoka, M.,
and
S. Kwano.
Mechanism of increased amplitude, and duration of the plateau with sudden shortening of diastolic intervals in rabbit ventricular cells.
Circ. Res.
54:
157-62,
1987[Abstract/Free Full Text].
19.
Hiraoka, M.,
A. Sunami,
Z. Fan,
and
T. Sawanobori.
Multiple ionic mechanisms of early afterdepolarizations in isolated ventricular myocytes from guinea-pig hearts. QT prolongation and ventricular arrhythmias.
Ann. NY Acad. Sci.
644:
33-47,
1992[Medline].
20.
Hondeghem, L. M.,
and
B. G. Katzung.
Antiarrhythmic agents: the modulated receptor mechanism of action of sodium and calcium channel-blocking drugs.
Annu. Rev. Pharmacol. Toxicol.
24:
387-423,
1884[Medline].
21.
Hondeghem, L. M.,
and
D. J. Snyders.
Class III antiarrhythmic agents have a lot of potential but a long way to go. Reduced effectiveness and dangers of reverse use dependence.
Circulation
81:
686-690,
1990[Abstract/Free Full Text].
22.
January, C. T.,
and
J. M. Riddle.
Early afterdepolarizations: mechanism of induction and block, a role for L-type Ca2+ current.
Circ. Res.
64:
977-990,
1989[Abstract/Free Full Text].
23.
Jurkiewicz, N. K.,
and
M. C. Sanguinetti.
Rate dependent prolongation of cardiac action potentials by a methanesulfonanilide class III antiarrhythmic agent: specific block of rapidly activating delayed rectifier K+ current by dofetilide.
Circ. Res.
72:
75-83,
1993[Abstract/Free Full Text].
24.
Kirsch, G. E.,
and
J. S. Sykes.
Temperature dependence of Na currents in rabbit and frog muscle membrane.
J. Gen. Physiol.
89:
239-251,
1987[Abstract/Free Full Text].
25.
Kiyosue, T.,
and
M. Arita.
Late sodium current and its contribution to action potential configuration of guinea-pig ventricular myocytes.
Circ. Res.
64:
389-397,
1989[Abstract/Free Full Text].
26.
Kiyosue, T.,
M. Arita,
H. Muramatsu,
A. J. Spindler,
and
D. Noble.
Ionic mechanisms of action potential prolongation at low temperature in guinea-pig ventricular myocytes.
J. Physiol. (Lond.)
468:
85-106,
1993[Abstract/Free Full Text].
27.
Kline, R. P.,
and
M. Morad.
Potassium efflux in heart muscle during activity: extracellular accumulation, and its implications.
J. Physiol. (Lond.)
280:
537-538,
1978[Abstract/Free Full Text].
28.
Levine, J. H.,
J. F. Spear,
T. Guarnieri,
M. L. Weisfeldt,
C. D. J. Delangen,
L. C. Becker,
and
E. N. Moore.
Cesium chloride induced long QT syndrome: demonstration of afterdepolarizations and triggered activity in vivo.
Circulation
72:
1092-1103,
1985[Abstract/Free Full Text].
29.
Roden, D. M.,
and
B. F. Hoffman.
Action potential prolongation and induction of abnormal automaticity by low quinidine concentrations in canine Purkinje fibers. Relationship to potassium and cycle length.
Circ. Res.
56:
857-867,
1986[Abstract/Free Full Text].
30.
Rosen, M. R.,
and
the Task Force of the Working Group on Arrhythmias of the European Society of Cardiology.
The Sicilian gambit: a new approach to the classification of antiarrhythmic drugs based on their actions on arrhythmogenic mechanisms.
Circulation
84:
1831-1851,
1991[Abstract/Free Full Text].
31.
Ruiz-Petrich, E.,
and
N. Leblanc.
The mechanism of the rate dependent changes of the conducted action potential in rabbit ventricle.
Can. J. Physiol. Pharmacol.
67:
780-787,
1989[Medline].
32.
Spear, J. S.,
B. G. Hook,
M. E. Josephson,
and
E. N. Moore.
Modulation of procainamide's effect on conduction by cellular uncoupling in perfused rabbit hearts.
J. Cardiovasc. Electrophysiol.
8:
199-214,
1997[Medline].
33.
Vassalle, M.
Electrogenic suppression of automaticity in sheep and dog Purkinje fibers.
Circ. Res.
27:
361-77,
1970[Abstract/Free Full Text].
34.
Yao, J.-A.,
E. J. Trybulski,
and
G.-N. Tseng.
Quinidine preferentially blocks the slow delayed rectifier potassium channel in the resting state.
J. Pharmacol. Exp. Ther.
279:
856-864,
1996[Abstract/Free Full Text].
AJP Heart Circ Physiol 274(3):H817-H828
0363-6135/98 $5.00
Copyright © 1998 the American Physiological Society
Top
Abstract
Introduction
