A Mechanism of Torsades de Pointes
in a Canine Model
GUST H. BARDY, M.D., Ross M. UNGERLEIDER, M.D., WILLIAM M. SMITH, PH.D.,
AND RAYMOND E. IDEKER, M.D., PH.D.
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
SUMMARY A dog model of torsades de pointes (TdP) was developed. Twenty 18-30-kg dogs had cardiopulmonary bypass instituted to maintain stable temperature, perfusion pressure and oxygenation. Quinidine, 30 mg/kg, was then administered and burst ventricular pacing was used to induce arrhythmias. The
left anterior descending coronary artery was occluded for 15 minutes and repeat pacing studies were
performed. Maps of epicardial activation were made from 27 simultaneously recorded electrograms obtained from 1-mm bipolar electrodes secured to the epicardium with a nylon mesh sock. Arrhythmias in five
dogs met criteria for the diagnosis of TdP: All had the characteristic undulating QRS morphology typically
associated with TdP, all occurred in the setting of QT prolongation and all ended spontaneously. The
epicardial maps demonstrated that each change in QRS morphology was associated with a change in the site
of epicardial breakthrough. Those QRS complexes during the transition from one morphology to the next
were associated with fusion cycles in which both the old and new sites of epicardial breakthrough were
present. In essence, two or more competing activation sequences were vying for control of epicardial
depolarization. This conclusion was strengthened by our ability to simulate TdP in the surface ECG and in
epicardial maps by simultaneously pacing from two widely separated ventricular sites at slightly different,
varying rates.
larity in an undulating fashion; it had to occur in the
setting of QT prolongation; and it had to terminate
spontaneously. Five dogs yielded 19 runs of TdP that
satisfied these criteria.
Figure 1 is a diagram of the experimental procedure.
TdP induction was attempted in 20 mongrel dogs that
weighed 18-30 kg. The dogs were anesthetized with
sodium pentobarbital, 30 mg/kg, and ventilated
through a Harvard respiration pump (model 607). A
median sternotomy was performed and the heart was
suspended in a pericardial cradle. Mean arterial pressure was monitored from the left femoral artery with a
Gould P23 ID Strantham pressure transducer through a
Hewlett Packard pressure amplifier (model 805C) and
viewed on a Tektronix Oscilloscope Display Screen
(model 7613). The left anterior descending coronary
artery (LAD) was loosely encircled with a 4-0 silk
ligature immediately distal to the first anterolateral
branch for subsequent occlusion. Because TdP is highly prone to degenerating into ventricular fibrillation,
cardiopulmonary bypass was instituted to help stabilize already precariously compromised dogs. Cardiopulmonary bypass was established with a bypass pump
(Sans Incorporated) and a Shiley IOOA blood oxygenator. Venous return to the oxygenator was through a
Bardic 38-gauge cannula inserted into the right atrium.
Blood was returned to the dog through a 16-gauge
cannula in the right femoral artery.
Epicardial recordings were made from 26 bipolar
electrodes separated from each other by 1 mm and
secured to the epicardium in a nylon mesh sock. 17 A
bipolar epicardial reference electrode was sutured to
the left ventricular epicardium. To ensure a normal,
stable core cardiac temperature throughout the experiment, myocardial temperature was monitored with an
Instrumentation Laboratory thermistor (catalog 44910)
inserted into the interventricular septum. Temperature
was adjusted with the bypass pump as necessary. Rapid ventricular pacing was performed from three sites
with bipolar hook electrodes constructed from Teflon-
TORSADES DE POINTES (TdP) is a ventricular
arrhythmia seen in the setting of QT prolongation.'-'
It has been associated with a wide array of clinical
entities, including congenital QT prolongation,5 6
atrioventricular block,3-5, 7quinidine intoxication,8 hypokalemia3-5 9 10 and myocardial infarction. "I Morphologically, it is characterized by an undulating QRS
with peaks that appear to rotate gradually around an
isoelectric baseline, as if the entire run were housed in
a sinusoidal envelope. 12 TdP is also known for its pernicious nature and for its responsiveness to isoproterenoll' 3' 5' 13 and to overdrive suppression pacing. l 3-5 14
The underlying electrophysiology of TdP is not
known. When Dessertenne originally described the arrhythmia in 1966 in an 80-year-old female with complete atrioventricular block, he favored the theory that
two competitive automatic ventricular pacemakers
were alternately controlling the heart.'2 Since then,
others have advanced different explanations, but none
has provided definitive evidence of the epicardial activation sequence underlying TdP. We developed a canine model of TdP to better define its epicardial activation sequence in a controlled laboratory setting with
the aid of epicardial mapping.
Methods
A ventricular arrhythmia was diagnosed as TdP if it
satisfied thtee criteria:'5 16 It had to be composed of
groups of QRS complexes that gradually changed poFrom the Department of Medicine, Division of Cardiology, and the
Departments of Pathology and Surgery, Duke University Medical Center, Durham, North Carolina.
Supported in part by research grants HL-07101, National Research
Service Award, and HL-17670, NHLBI.
Dr. Ideker is the recipient of Research Career Development Award
HL-00546.
Address for correspondence: Gust H. Bardy, M.D., Box 31121,
Duke University Medical Center, Durham, North Carolina 27710.
Received January 25, 1982; revision accepted July 6, 1982.
Circulation 67, No. 1, 1983.
52
TORSADES DE POINTES/Bardy et al.
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
FIGURE 1. Experimental procedure. Cardiopulmonary bypass was established by oxygenating right atrial blood and
returning it through the rightfemoral artery. Cardiac temperature could be stabilized with the pump oxygenator to 37-38°C.
Arterial pressure was monitored from the left femoral artery
and stabilized with the pump oxygenator to a mean of 60-80 mm
Hg. After oxygenation, temperature and pressure were stabilized, quinidine was administered, followed by left anterior descending coronary artery (LAD) occlusion. Arrhythmias were
then induced by rapid ventricular pacing from one of the bipolar
pacing electrodes inserted into the ventricular endocardium.
coated silver wires and plunged through the ventricles
with a 19-gauge needle, which was then withdrawn,
leaving the silver wires lodged against the endocardium. Electrograms from limb leads I, II and III were
then simultaneously recorded at a paper speed of 100
mm/sec and baseline QT intervals were measured as
the interval between the earliest QRS deflection and
the terminal part of the T wave from the limb lead
demonstrating the longest QT interval. Quinidine was
administered over 15 minutes at a loading dose of 30
mg/kg, followed by a constant infusion of 5.6 ,ug/kg/
min. This dose was chosen to produce toxic quinidine
levels according to known pharmacokinetics modified
for expected changes in the volume of distribution as a
result of cardiopulmonary bypass.'>20 Blood samples
were obtained at the end of the study to determine
quinidine levels by photofluorometric precipitation.
Burst pacing from the right ventricular outflow tract,
the right ventricular apex and the left ventricular free
wall near the base was done with eight-beat salvos
delivered with a stimulus strength twice diastolic
threshold. Stimuli were initially delivered with a cycle
length of 250 msec and then decreased in 5-msec decrements until a rapid ventricular arrhythmia resulted or
the pacing stimulus failed to capture the ventricle on a
one to one basis. If no ventricular arrhythmia could be
induced from any of the three pacing sites, the LAD
was occluded; after 15 minutes, the pacing protocol
was repeated. At the end of the study, QT intervals
were again recorded as described above.
Electrocardiographic limb leads I, II and III and 27
bipolar epicardial signals were recorded continuously
on a 32-channel FM analog tape recorder (Ampex PR220). The epicardial potentials recorded during ven-
53
tricular arrhythmias were played back from the tape
recorder, digitized and entered in a computer (Digital
Equipment Corporation PDP- 1 1/34) as described elsewhere.2' Local activation times at each electrode were
determined for each ventricular cycle. The activation
times for each bipolar electrogram are preliminarily
chosen by the computer and displayed on a Tektronix
4014 graphics terminal. All computer-chosen activation times are then reviewed by the investigator for
accuracy. Adjustments can be made by hand if necessary. Once the position of each bipolar electrode is
designated on the epicardial map for the computer,
isochrones of epicardial activation can be constructed
by the computer using previously chosen local activation times. Isochronous maps of epicardial activation
can then be displayed on the Tektronix graphics terminal. Details of this analysis have been described.22
In two dogs, TdP was simulated by pacing at different rates from two widely separated endocardial sites.
Stimuli were delivered through bipolar plunge electrodes. One electrode was secured at the right ventricular endocardium and the other at the left ventricular
endocardium. Each site was paced with a stimulus
strength of 2 mA and a pulse width of 4 msec. The
right ventricular site was paced at a constant cycle
length of 180 msec. Cycle length at the left ventricular
site was gradually changed from 165 to 195 msesc and
back to 165 msec.
Results
of
Five the 20 dogs yielded runs of a ventricular
arrhythmia that met the criteria for TdP. From these
five studies, 19 runs of TdP were recorded and analyzed. All five dogs had prolonged QT intervals after
quinidine compared with baseline recordings (table 1).
The mean QT interval was 0.21 + 0.01 second (±+SD)
before and 0.27 + 0.04 second after quinidine administration (p = 0.0 16 by paired t test). The mean quinidine level in these five dogs was 6.4 mg/l. The normal
QT interval for dogs has been reported as 0.177 ±
0.054 second23 and 0.21 + 02 second.24 The duration
of TdP ranged from 2.7 to 160 seconds (12-820 cycles) (mean 43 + 46 seconds). All runs terminated
spontaneously. TdP was induced in one dog with
quinidine alone (quinidine level of 4.6 mg/l), but the
TABLE 1. QT-interval Changes After Quinidine Administration
and Left Anterior Descending Coronary Artery Occlusion in Five
Dogs with Torsades de Pointes
QT interval
Quinidine
After
level
Before
(sec)
(sec)
(mg/l)
0.23
4.6
0.21
0.21
0.25
5.8
0.22
0.31
5.9
7.8
0.25
0.20
0.31
8.1
0.21
6.4
0.27
0.21
Mean
Before = before quinidine or occlusion; After = after quinidine
and occlusion.
VOL 67, No 1, JANUARY 1983
CIRCULATION
54
B
APEX
SEX
UMS LEAon
UMS LEADJ
MtvtMJYCNVnvrVr\P\
MJPJtPJA
+
I SI
fwVvmAn~~rvvw\\4di~ ~ ~ RPPPA/
~ J-,
v
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
APEX
AEX
UMS LEADUL
UMB LEADn
fJAAJA\r0J
+I
FIGURE 2. Correlation of isochronous maps of epicardial activation with changes in QRS morphology during torsades de pointes.
Every other cycle is presented through a typical transition from one QRS morphology to another and back to the first again. Each panel
demonstrates the map of the cycle identified by the arrow in limb lead II. The heart is presented as ifcutfrom apex to crux along the
posterior interventricular groove and then flattened. Each number represents local activation time (msec) relative to the point of
earliest epicardial breakthrough, which is indicated by zero. Isochrones are shown every 10 msec and estimate the location of the
depolarization front at one instant during excitation. The asterisks denote the location of recording electrodes over ischemic
myocardium that failed to register discrete epicardial activation. (A) Epicardial activation is primarily controlled from a site that
breaks through on the basilar anterior left ventricular surface. A second, relatively early site is at the apex 41 msec after earliest
epicardial breakthrough. (B) Epicardial activation undergoes subtle changes near the inferior margin of the infarct. (C) Transition
complex. Simultaneous epicardial breakthrough occurs from the basilar and apical left ventricle (LV). Note the associated change in
the QRS morphology in limb lead II. (D) The apical location captures complete control of epicardial depolarization, with the high
anterior LV no longer demonstrating early breakthrough. (E) There is a gradual lengthening in total duration of epicardial activation,
but the pattern of activation is generally unchanged. (F) The site of earliest epicardial breakthrough shifts slightlyfrom the apex to the
posterior LV, with a small change in the QRS morphology as seen in limb lead II. (G) The basilar anterior LV site of epicardial
breakthrough reappears near simultaneously with the apically located site of earliest breakthrough. At this point, QRS morphology
dramatically changes and appears to be transitional between the two basic morphologies. (H) The basilar anterior LV site resumes
predominant contol of epicardial depolarization and the limb lead QRS is upright again. RV = right ventricle; PA = pulmonary
artery; LAD = left anterior descending coronary artery.
other four dogs required LAD occlusion as well as
quinidine before TdP could be induced. In eight runs, a
typically uniform run of ventricular tachycardia was
intermingled with TdP.
Of the 15 dogs that failed to manifest TdP, 12 had a
prolonged QT interval of at least 20% over baseline.
These 15 dogs without TdP did, however, yield mono-
morphic ventricular tachycardia, ventricular fibrillation or both. All runs of monomorphic ventricular
tachycardia manifested only one site of epicardial
breakthrough during epicardial activation mapping.
We do not know why these dogs did not develop TdP.
Isochronous maps of TdP revealed that epicardial
activation occurred in discrete cycles and that each
TORSADES DE POINTES/Bardy et al.
55
APEX
UMS LADn
UMSLEAOD
e
wi
IK
4
AMW/atW
A ~
l~
1%.IA
l!e1J%I-AI,
4~~~~o
H
G
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
PEKX
APEX
LIMB LEADII
UMS LEADfl
d4
500
I MI
QRS complex was associated with a discrete epicardial
activation sequence. In each run of TdP, the QRS
morphology changed in concert with a change in the
site of earliest epicardial activation and in the pattern
of epicardial spread. The degree of change in QRS
morphology followed the degree of change in epicardial activation. If the site of epicardial breakthrough
moved to a nearby site, then the QRS morphology
changed only slightly (figs. 2E and F). If the earliest
site of epicardial activation moved to a considerably
different area of the epicardium, then the QRS mor-
phology changed more dramatically (figs. 2F and H).
A shift from an apical location to a basal location for
the site of earliest epicardial activation caused the QRS
polarity in limb lead II to change from negative to
positive.
Two to seven distinct sites of epicardial breakthrough were seen during TdP. Nine runs of TdP had
two foci of epicardial breakthrough, five had three
foci, three had four foci, one run had five foci, and one
had seven foci (fig. 3). During the transition from one
QRS morphology to another, changes in the epicardial
FIGuRE 3. A limb lead II recording of a run
of torsades de pointes induced with an eightbeat salvo of rapid ventricular pacing (cycle
-04-
1000 nec
length 165 msec). The arrow designates the
last paced beat. The three strips are continuous. The highly varying QRS morphology and
cycle length are manifestations of seven distinct sites of earliest epicardial breakthrough
alternating control of epicardial activation
during this run of torsades de pointes. Some
QRS morphologies are seen repeatedly.
VOL 67, No 1, JANUARY 1983
CIRCULATION
56
B
LIMB LEADIE
LIMB LEADII
+
I
+
cI
I
c
I
C
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
APEX
APEX
LIMB LEAD1
LIMB LEADII
>X 500
rnmsec
APEX
LIMB
FIGURE 4. Correlation of isochronous maps ofepicardial activation with changing QRS morphology during simulation of
torsades de pointes by two widely separated endocardial pacemakers firing at varying rates. Stimuli are delivered to bipolar
plunge electrodes secured to the right and left ventricular (RV
and LV) endocardium. The format is similar to that offigure 2.
(A) The LV pacing site predominates with early epicardial
breakthrough occurring high on the anterior LV near the left
anterior descending coronary artery (LAD). (E) Control of
epicardial activation gradually shifts to the RV pacing site
which demonstrates earliest epicardial breakthrough near the
RVapex. (I) Epicardial activation is again controlled by the LV
pacing site and is nearly the same as the epicardial activation
pattern in panel A.
LEAD
site of breakthrough could be abrupt. Usually, howevthe transitional QRS complexes were associated
with epicardial maps that demonstrated two separate
sites of epicardial breakthrough (figs. 2C and G). During the transition, one dominant breakthrough site
gradually yielded to the other (fig. 2). Furthermore,
er,
X~~~~~~
Isoc
the QRS width of the transition complexes were narrower than nontransition complexes as a result of wave
front fusion.
Cycle lengths for each morphologic grouping of
complexes within a run of TdP were highly variable,
ranging from 118 to 322 msec (mean 186 + 36 msec).
The cycle lengths were similar within a morphologically similar group of QRS complexes. However, cycle
length would change after a change in the site of epicardial breakthrough which, in turn, resulted in a dif-
TORSADES DE POINTES/Bardy et al.
57
G
APEX
LIO LEADII
LIMB LEADH
~~I
4
~~~~~~~~~~~~~I^u
SOO6 I
I
e
SOO
sec
I
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
APEX
APEX
LIMB LEADE
LIMB LEADI
'S~~~~~~~M
ferent QRS morphology. There was no pattern in the
variation in cycle lengths. The QRS complexes with
shorter cycle lengths did not necessarily predominate.
Some epicardial foci of earliest activation with shorter
cycle lengths were superceded by other epicardial foci
of earliest activation with longer cycle lengths.
In the two dogs in which it was attempted, TdP was
simulated by pacing at varying rates from two widely
separated endocardial sites. As in the TdP observed
after quinidine administration and acute myocardial
ischemia, changes in QRS morphology correlated
closely with changes in the site of earliest epicardial
activation (fig. 4). All transitional QRS complexes
were associated with epicardial maps that showed fusion of epicardial activation as it spread from the two
sites of epicardial breakthrough (fig. 4C). The transition QRS complexes during simulated TdP were also
narrower than nontransitional complexes, a finding
consistent with fusion cycles and shorter activation
times.
Discussion
Little progress has been made in defining the genesis
of TdP. Some investigators have implicated reentry7' 25
as its basic mechanism. Dessertenne'2 favored an automatic mechanism. To resolve the reentry-automaticity
controversy, mapping studies with a large number of
endocardial, intramyocardial and epicardial electrodes
are required. However, mapping with a moderate
number of electrodes distributed over the epicardium
explains one aspect of the electrophysiology of TdP:
how TdP creates the characteristic QRS morphologic
changes in the surface ECG for which it is distinctive.
Several mechanisms have been proposed to explain
the morphologic features of TdP, but none has been
substantiated. Desertenne'2 believed that the characteristic electrocardiographic findings may occur if two
automatic ventricular pacemakers, firing slightly out
of phase, alternate control of the ventricles. Smirk and
Ng26 envisoned TdP as an idioventricular focus intermittently subject to progressive ventricular refractoriness, as if spread of activation were encountering a
global ventricular Wenckebach phenomenon. Ahronheim27 suggested that TdP results from a precession of
the mean QRS axis. He proposed that during a tachyarrhythmia, if the ventricular polarization cycle is destabilized, the mean QRS vector could gyrate in space.
This study of epicardial activation during TdP does
58
CIRCULATION
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
establish one explanation for the electrocardiographic
features of TdP. The QRS morphology of the surface
ECG and the sequence of epicardial activation are
closely linked: The unique morphologic characteristic
of TdP is apparently related to a changing pattern of
epicardial breakthrough and epicardial spread (fig. 2).
There was no evidence to support a mechanism of a
global ventricular refractoriness or global Wenckebach
phenomenon. Furthermore, a precession of QRS axis
was not seen. Similarly, the change in QRS morphology was not associated with acceleration or deceleration of the epicardial activation wave front arising from
a single focus.
The findings that QRS morphology changes in concert with a change in the site of earliest epicardial
activation and with a change in the pattern of epicardial
spread were verified when TdP was simulated. Simulated TdP was accomplished by pacing simultaneously
from two widely separated endocardial sites in two
dogs that did not undergo any surgical or pharmacologic interventions. With one pacing wire in the right
ventricular endocardium and the other in the left, simultaneous pacing at different and varying rates yielded a QRS that alternated from a right to a left bundle
branch block pattern according to which pacing site
primarily controlled epicardial activation and the site
of earliest epicardial breakthrough (fig. 4). This simulated TdP had the same morphologic characteristics of
that in quinidine-intoxicated, infarcted dogs; that is,
simulated TdP showed alternating foci of early epicardial breakthrough and changing patterns of epicardial
spread of activation as well as narrow transition
complexes.
These observations on the electrocardiographic features of TdP as they relate to epicardial activation are
founded on a model that, in part, differs from clinical
TdP. It is a complex model that incorporates quinidine
administration with coronary occlusion and ventricular
pacing in the presence of cardiopulmonary bypass.
Although TdP certainly occurs clinically during quinidine administration with resultant QT interval prolongation8'230 and has recently been reported during
myocardial infarction," it is a spontaneous rhythm.
Our model differs from the clinical situation in that
burst pacing was necessary to induce TdP in four of
five dogs. Furthermore, a critical aspect of our protocol was cardiopulmonary bypass. In our experience,
open-chested, acutely infarcted dogs are prone to fluctuation in core temperature, blood pressure and cardiac
output. The latter two variables can be particularly
unstable during multiple runs of ventricular arrhythmias. By stabilizing temperature and hemodynamic
variables with cardiopulmonary bypass, ventricular arrhythmias were much more easily sustained and reproducible (Bardy GH et al.: unpublished observations).
The fact that some runs of TdP exceeded 160 seconds
may in part be attributed to hemodynamic stability
provided by cardiopulmonary bypass.
Given that this model has extenuating aspects to it, it
nevertheless yielded 19 tachyarrhythmias that by all
known electrocardiographic criteria satisfied the defi-
VOL 67, No 1, JANUARY 1983
nition of TdP. The QRS morphology characteristically
changed in an undulating fashion; the arrhythmia terminated spontaneously, distinguishing it from ventricular fibrillation; and the QT interval was prolonged.
While it may be hypothesized that TdP occurs by other
mechanisms in other situations, this remains to be
shown.
Crucial to understanding TdP is whether reentry or
automaticity occurs in the localized region that encompasses the arrhythmogenic tissue. However, in our
model, this question is not the key determinant for the
unusual electrocardiographic finding that has piqued
such interest in this arrhythmia. Whatever mechanism
gives rise to the different epicardial activation sequences with their different sites of epicardial breakthrough, the QRS morphology is still determined by
spread of activation away from the localized region of
arrhythmia origin throughout the remaining ventricular
muscle.
In our model, observations in simulated and in induced TdP indicate that TdP can be characterized as a
ventricular arrhythmia with two or more epicardial
breakthrough sites vying for control of epicardial activation. With each change in the earliest site of epicardial activation, a change in QRS morphology can be
expected. The narrow transition complexes, or "spindles" that link morphologically distinct QRS complexes, result from fusion of two colliding cycles of
epicardial depolarization.
Acknowledgment
We are indebted to Marjorie Penny for invaluable secretarial assistance and to Dave Huggett and Francine Mehler for preparing the
illustrations. We also thank Jimmy Manley for knowledgeable technical
assistance.
References
1. Krikler DM, Cunry PVL: Torsade de pointes, an atypical ventricular tachycardia. Br Heart J 38: 117, 1976
2. Mallion J-M, Aveyou F, Denis B, Yaccoub M, Martin-Noel P:
Syndrome de QT long over torsades de pointe, syncopes et insuffisance coronarienne. Arch Mal Coeur 65: 1209, 1972
3. Motte G, Coumel PH, Abitol G, Dessertenne F, Slama R: Le
syndrome QT long et syncopes par "torsades de pointe." Arch
Mal Coeur 63: 831, 1970
4. Slama R, Motte G, Coumel PH, Dessertenne F: Le syndrome
"allongement de QT et syncopes par torsades de pointes". Lavel
Med 42: 353, 1971
5. Slama R, Coumel PH, Motte G, Gourgon R, Waynberger M,
Touche S: Tachycardies ventriculaires et torsades de pointes. Frontieres morphologiques entre les dysrythmies ventriculaires. Arch
Mal Coeur 66: 1401, 1973
6. Schwartz PJ, Malliani A: Experimental and laboratory reports:
electrical alternation of the T-wave: clinical and experimental evidence of its relationship with the sympathetic nervous system and
with the long Q-T syndrome. Am Heart J 89: 45, 1975
7. Fontaine GH, Frank R, Welti JJ, Grosgogeat Y: Electophysiology
of torsades de pointes. In Proceedings of the VIth World Symposium on Cardiac Pacing, edited by Meere C. Montreal, Pacesym,
1979, ch 6
8. Reynolds EW, Vander Ark CR: Quinidine syncope and the delayed
repolarization syndromes. Mod Concepts Cardiovasc Dis 45: 117,
1976
9. Curry P, Fitchett D, Stubbs W, Krikler D: Ventricular arrhythmias
and hypokalemia. Lancet 2: 231, 1976
10. Raynaud Brochier M, Neel JL, Faucher JP, Raynaud PH: Tachycardia ventriculaire A foyer variable et dyskalieme. Arch Mal Coeur
TORSADES DE POINTES/Bardy et al.
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
62: 1578, 1969
11. Doroghazi RM, Childers R: Time related changes in the Q-T interval in acute myocardial infarction: possible relation to local hypocalcemia. Am J Cardiol 41: 684, 1978
12. Dessertenne F: La tachycardie ventriculaire a deux foyers opposes
variable. Arch Mal Coeur 59: 263, 1966
13. Cokkinos DV, Mallios C, Philias N, Vorides M: "Torsade de
pointe": a distinct entity of ventricular arrhythmia? Acta Cardiol
(Brux) 33: 167, 1978
14. Brochier M, Fauchier JP: Torsades de pointe et rentr6es provoqu6es par les antiarythmiques. Arch Mal Coeur 71: 477, 1978
15. Kossmann CE: Torsade de pointes: an addition to the nosography
of ventricular tachycardia. Am J Cardiol 42: 1054, 1978
16. Smith WM, Gallagher JJ: "Les torsades de pointes": an unusual
ventricular arrhythmia. Ann Intern Med 93: 578, 1980
17. Harrison L, Ideker RE, Smith WM, Klein GJ, Kasell J, Wallace
AG, Gallagher JJ: The sock electrode array: a tool for determining
global epicardial activation during unstable arrhythmias. PACE 3:
531, 1980
18. Greenblatt DJ, Pfiefer HJ, Ochs HR, Franke K, MacLaughlin DS,
Smith TW, Koch-Weser J: Pharmacokinetics of quinidine in humans after intravenous, intramuscular and oral administration. J
Pharmacol Exp Ther 202: 365, 1977
19. Wallace AG, Cline RE, Sealy WC, Young WG, Troyer WG:
Electrophysiologic effects of quinidine. Studies using chronically
implanted electrodes in awake dogs with and without cardiac denervation. Circ Res 19: 960, 1966
20. Ueda Ct, Hirschfeld DS, Scheinman MM, Rowland M, William-
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
59
son BJ, Dzindzio BS: Disposition kinetics of quinidine. Clin Pharmacol Ther 19: 30, 1976
Ideker RE, Smith WM, Wallace AG, Kasel J, Harrison L, Klein
GJ, Kinicki RE, Gallagher JJ: A computerized method for the rapid
display of ventricular activation during the intraoperative study of
arrhythmias. Circulation 59: 449, 1979
Smith WM, Ideker RE, Kinicki RE, Harrison L: A computer system for the intraoperative mapping of ventricular arrhythmias.
Comput Biomed Res 13: 61, 1980
Burnan SO, Kahn S, Panagopoulos P: The electrocardiogram of
the normal dog. J Thorac Cardiovasc Surg 51: 379, 1966
Horowitz SA, Spanier MR, Wiggers HC: The electrocardiogram of
the normal dog. Proc Soc Exp Biol Med 84: 121, 1953
Horowitz LN, Greenspan AM, Spielman SR, Josephson ME: Torsade de pointes: electrophysiologic studies in patients without transient pharmacologic or metabolic abnormalities. Circulation 63:
1120, 1981
Smirk FH, Ng J: Cardiac ballet: repetition of complex electrocardiographic patterns. Br Heart J 38: 117, 1969
Ahronheim GA: Letter to the editor. N Engl J Med 94: 277, 1981
Selzer A, Wray HW: Quinidine syncope. Paroxysmal ventricular
fibrillation occurring during treatment of chronic atrial arrhythmias. Circulation 30: 17, 1964
Koster RW, Wellens HJJ: Quinidine-induced ventricular flutter
and fibrillation without digitalis therapy. Am J Cardiol 38: 519,
1976
Schwartz PJ, Periti M, Malliani A: The long QT syndrome. Am
Heart J 89: 378, 1975
A mechanism of torsades de pointes in a canine model.
G H Bardy, R M Ungerleider, W M Smith and R E Ideker
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
Circulation. 1983;67:52-59
doi: 10.1161/01.CIR.67.1.52
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1983 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://circ.ahajournals.org/content/67/1/52
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Circulation 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 is online at:
http://circ.ahajournals.org//subscriptions/