Experimental models of heart failure

European Heart Journal Supplements (2004) 6 (Supplement F), F7–F15
Experimental models of heart failure
Emilio Vanolia,b,c,*, Sara Bacchinia, Stefania Panigadaa,
Francesco Pentimallia, Philip B Adamsonb,c
Department of Cardiology, IRCCS Policlinico S. Matteo and University of Pavia, Pavia, Italy
Departments of Physiology, Oklahoma University Health Sciences Center, Oklahoma City, OK, USA
Department of Medicine Cardiology, Oklahoma University Health Sciences Center, Oklahoma City, OK, USA
Despite significant progress in the prevention and treatment of cardiovascular diseases,1 the incidence and
prevalence of congestive heart failure (CHF) have been
* Correspondence: Emilio Vanoli, MD, FESC, Department of Cardiology,
IRCCS San Matteo and University of Pavia, Italy. Tel./fax: +39 382529531.
E-mail address: [email protected] (E. Vanoli).
increasing steadily in recent years,2 especially in the elderly.3 The most common cause of chronic heart failure is
no longer hypertension or valvular heart disease, as it
was in past decades, but rather coronary artery disease
(CAD).4 Heart Failure is the consequence of multiple
pathophysiological alterations and adaptations, leading
to left ventricular (LV) hypertrophy, dysfunction and dilatation, increased systemic vascular resistance and activation of the neuroendocrine system.5 This latter seems
to be to most critical component of all as the only
1520-765X/$ - see front matter c 2004 The European Society of Cardiology. Published by Elsevier Ltd. All rights reserved.
Downloaded from by guest on December 3, 2014
Background Most successful and reproducible animal models of chronic heart failure
(HF) include: increased peripheral resistance, rapid ventricular pacing, arteriovenous
shunt, supravalvular stenosis, chronic heart block, repetitive transmyocardial direct
current (DC) shock. The main limitation of all these approaches was the lack of stability of the left ventricular (LV) dysfunction.
Methods We have described a stable and reproducible animal model of LV dysfunction
mediated by ischemic loss of contractive myocardium, which is suitable for chronically studying the complex pathophysiology of HF. This model incorporates conscious
dogs with a chronic anterior myocardial infarction (MI). Once the MI is stabilized the
progression of the ischemic disease is produced by injecting into the circumflex coronary artery 1–2 cc of latex microspheres to cause embolization of microcirculation.
This procedure is repeated 3–5 times over several weeks and, at the end, it results in
a spread ischemic damage and a significant loss of ventricular systolic function.
Results The healed MI results in a depression of heart rate variability and baroreflex
sensitivity in a subgroup of dogs. In these dogs the depressed vagal control of the
heart is associated with elevated arrhythmic risk. On the contrary in dogs with preserved cardiac reflexes the arrhythmic risk is low. In the high risk group a profound
electrophysiological remodelling occurs with the MI and progresses once LV dysfunction is created while the low risk dogs tolerate the embolization procedure with a loss
in LV function but without the changes in the cardiac electrical stability.
Conclusion Experimental preparations based on a chronic MI and progression of the
ischemic damage toward a chronic LV dysfunction has provided important information
concerning autonomic and electrophysiologic alterations associated with sudden cardiac death.
c 2004 The European Society of Cardiology. Published by Elsevier Ltd. All rights
Heart failure
E. Vanoli et al.
physiologic alterations leading to sudden arrhythmic
Experimental approaches to heart failure
Experimental preparations of HF have involved various
interventions including trauma12 and toxic depression
of cardiac function.13 However, these interventions
may directly affect neuro-humoral factors independently
from the primarily influence of HF, while the experimental preparation should allow quantifiable results that can
be extrapolated to the clinical reality of the disease.
Others approaches to the creation of an animal model
of HF were more successful and reproducible. Such models include increased peripheral resistance,14 rapid ventricular pacing,15 creation of an arteriovenous shunt,16
supravalvular stenosis,17 chronic heart block,18 repetitive transmyocardial direct current (DC) shock,19 but,
again, very few of these models generated data that
could be consistently translated into human applications.
Tachycardia-induced cardiomyopathy
Among the many approaches, the tachycardia induced
model has been one of the most used. Sympathetic
dependent increased heart rate is a typical initial mani-
Table 1
Naturally occurring models
Dilated cardiomyopathy
Salt-sensitive hypertension
Hamstera, dog, turkey
Experimentally lnduced Models
Myocardial ischemia
Coronary ligation
Coronary embolism
Electrical shock
Rata, dog, pig, rabbit
Dog, pig
Chronic rapid cardiac pacing
Ventricular pacing
Supraventricular pacing
Doga, pig, rabbit
Doga, rabbit
Pressure overload
Aortic banding
Pulmonary artery banding
Rat, guinea pig
Mouse, rat, cat, dog, pig
Volume overload
Arteriovenous shunt
Mitral regurgitation
Aortic regurgitation
Rat, dog
Toxic cardiomyopathy
Rat, rabbit, dog, pig
Rat, turkey
Genetically altered animals
Dilated cardiomyopathy
Most frequently used models.
Downloaded from by guest on December 3, 2014
currently effective pharmacological therapies are b-adrenergic receptors blockers6,7 and inhibitors of the angiotensin converting enzyme8,9 while other interventions
targeted to other important factors of heart failure
(HF) such, for instance, TNF had failed to show any efficacy.10 The role of the various mechanisms responsible
for HF progression is difficult to ascertain in humans because of uncertainty associated with the identification of
when HF in fact begins. An added difficulty is the usual
confounding influence of concomitant pharmacological
therapies. The understanding of left ventricular remodelling and dysfunction is fundamental to describe the
natural history of the disease and the efficacy and timing
of interventions needed to positively interfere with all
the processes leading to HF and, ultimately, cardiac
death. In this context it is important to remember that
mortality in the early stages of HF is strongly determined
by arrhythmic mechanisms while in more advanced
stages pump failure is a more frequent mechanism. Because of these many difficulties, animal models of
chronic HF are fundamental to describe the complex nature of this disease process and a number of experimental
model systems have been generated in various species
(see Table 1).11 Hereafter, we will briefly review some
aspects of the many experimental models of heart failure
described in recent years. We will then describe a stable
and reproducible animal model of LV dysfunction mediated by ischemic loss of contractive myocardium, which
is suitable for studying the complex pathophysiology of
this disease process with a specific focus on the electro-
Experimental Models of Heart Failure
repolarizing currents explain the abnormal prolongation
of repolarization and its dispersion and describes a
potential arrhythmogenic mechanism in HF. Overall,
the tachycardia-induced HF models have provided a number of important information concerning cellular and
hemodynamic changes occurring in HF. Nonetheless, this
model suffers the limitation of been unstable. This feature has made impossible, for instance, a substantial
understanding about autonomic mechanisms in HF as
they can be derived from the analysis of heart rate. Additionally, fast heart rate is, now a days, an infrequent
cause of HF and, thus, some important aspects of this
syndrome may be overlooked by using this model. One
for all is the fact that the myocardial damage induced
by fast pacing is homogeneous and thus, those many aspects that are dependent upon the typical mechanical
and electrical dysomogeneity of the failing heart cannot
be assessed in this type of preparation. Based in these
considerations the attention of experimental investigators has focused on ischemic models of heart failure. This
considering that the ethyology of LV dysfunction and
heart failure is mostly ischemic.
Multiple coronary microembolizations
Initial attempts in the use of microembolization implied
a single intracoronary injection of microspheres to produce an extensive myocardial damage.31–34 This approach to cause a sustained LV dysfunction was, in
most cases, lethal and thus, not suitable for chronic
models. However, if the same amount of damage was
produced by intermediate steps so that partial recovery
was allowed between each ischemic insult, the changes
for a longer survival would have been higher. Based on
this idea, Sabbah et al.35 developed a stable model of
CHF performing multiple sequential intracoronary embolization with polystyrene latex microspheres. The initial
study was performed in 20 dogs in which 3–9 embolizations were performed 1 week apart. The first 3 embolizations consisted of 2 ml of microsphere suspension
injected subselectively into either the left anterior
descending or left circumflex coronary artery in an alternating fashion. Subsequent embolizations consisted of
3–6 ml of microspheres divided equally between the left
anterior descending or left circumflex coronary artery
until LV ejection fraction was <35%.
Among the 20 dogs studied six (30%) died before full
completion of the study. One dog died 24 h after the first
embolization from a rupture posterior wall infarction.
The remaining five deaths were due to anesthesia in
one dog, cardiac decompensation in two, and sudden
and unexpected cardiac death, presumably due to
arrhythmias in two dogs. The dogs developed a chronic
HF mediated by a loss of contractile myocardium and
manifested many of the sequelae of HF including marked
depression of LV systolic and diastolic function, LV dilatation and hypertrophy, reduced cardiac output, development of mitral regurgitation and elevation of
systemic vascular resistance. The depression of LV
Downloaded from by guest on December 3, 2014
festation of LV dysfunction evolving into HF and high
heart rate is indicated as a marker but also a mechanism
of the failing heart.
The effect of tachycardia on ventricular structure and
function is an important area of research because incessant or chronic elevated heart rate, as it may occur in patients with atrial fibrillation with elevated ventricular
responses not adequately modulated by the therapy, produces ventricular dysfunction and dilation. Experimental
tachycardia-induced cardiomyopathy was first described
by Whipple et al.20 in 1962. In 1986 Armstrong et al.16
examined rapid ventricular cardiac pacing as a means
of inducing HF in the dog. They instrumented 15 mongrel
dogs with a unipolar pacemaker lead placed in the right
ventricular apex. Seven animals were paced at 250
beats/min for 3 weeks (VP1 group), an additional series
of six animals (VP2 group) was paced until a clear biologic end point for HF was reached. The changes in cardiac size and hemodynamics in the VP1 and VP2 group
were compared with those obtained in parallel studies
of 10 sham-operated animals. The VP1 group showed an
increase in cardiac size, a reduction in mean arterial
pressure, a fall in cardiac index and an increase in left
ventricular filling pressure. The VP2 group showed similar
but more advanced changes. The changes in these two
groups were significantly different from those in the
group of the sham-operated animals. A neurohumoral
evaluation on the animals showed that plasma norepinephrine and renin activity were unchanged in the sham
operated animals, whereas in the VP1 group, plasma norepinephrine rose but plasma renin activity did not
change. In the VP2 group both norepinephrine and plasma renin rose. In this prospective the tachycardia model
reproduces some of the conditions associated with HF
and indeed, a large number of animal models21–27 including the one we have just described, support the concept
that incessant or chronic tachycardia can lead to severe
bi-ventricular systolic and diastolic dysfunction characterized by hemodynamic and cardiac structural changes.
These changes occur as soon as 24 h after rapid pacing,
with continued deterioration in ventricular function for
up to 3–5 weeks resulting in end-stage HF. The main limitation of this preparation is that all the mechanical
alterations induced by rapid pacing reverse in few days
after stopping the treatment. Such a recovery from pacing-induced cardiomyopathy demonstrates that the myopathic process associated with rapid heart rates is largely
reversible. Within 48 h after stopping pacing, hemodynamic variables approach control levels, and left ventricular ejection fraction shows significant recovery
with subsequent normalization after 1–2 weeks.
Although first devised to mimic tachycardia-induced
cardiomyopathy in humans, the model has been invaluable to the general study of heart failure by providing a
predictable, although unstable, model of low output
biventricular failure.16,21–29 Of specific importance is
the information that this model has provided on ionic
changes in HF. Marban and Tommaselli indeed documented repolarization abnormalities30 leading to sudden
death due to a significant reduction in the expression of
repolarizing currents, namely Ito and Ik1.31 This loss in
E. Vanoli et al.
1 Second
ejection fraction, does not impede the process of progressive chamber dilation. On the other hand, this experimental preparation has not provided major information
about the arrhythmias mechanisms responsible for sudden death in ischemic HF. This mainly because no risk
stratification for arrhythmic death has been described
in this preparation. The analysis of the arrhythmia pattern in this model.37 has shown that some dogs will develop ventricular tachyarrhythmias but no information
has emerged, so far, from studies in this model that
had enhanced the understanding of arrhythmic sudden
death in HF.
An experimental model of spontaneous
sudden death in chronic post-MI ischemic
LV dysfunction
The microembolization technique has been applied in a
canine model for post-MI sudden death described by
Schwartz and Stone. In this earlier model38 the interaction between acute myocardial ischemia and disturbances in the autonomic nervous system plays a key
role in sudden cardiac death, particularly when these
factors occur in a myocardium that is electrically unstable because of pre-existing ischemic damage.
The model consists of two stages. In the first stage an
anterior wall myocardial infarction (MI) is produced by a
permanent ligation of the anterior intraventricular
branch of the left coronary artery immediately proximal
to the first major diagonal artery perforator. During the
same surgical session a pneumatic occluder is placed
60 Sec CAO
Fig. 1 Continuous transthoracic lead I ECG demonstrating ventricular fibrillation in a susceptible dog during submaximal exercise and transient
myocardial ischemia characteristic of the model described in this study. The arrow denotes 60 s following inflation of the circumflex coronary artery
occluder and the point when the animal stopped exercising (see text for details). Note the abrupt onset of ventricular tachycardia degenerating quickly
into ventricular fibrillation (From Ref. 44).
Downloaded from by guest on December 3, 2014
function is accompanied, in this model, by activation of
the sympathetic nervous system and by increased secretion of ANF. An important feature of this model of HF is
the observation of a lack of recovery of LV function once
coronary embolization were discontinued. It is likely that
multiple embolization, repeated over time, gradually exausts the compensatory mechanisms to counteract the
loss of viable tissue and, therefore, leads to a sustained
depression of cardiac function. In the present model
plasma norepinephrine concentration increased substantially while plasma renine activity (PRA) remained within
normal limits throughout the course of evolving HF. This
was at variance from studies in tachycardia induced HF in
which both PRA and norepinephrine plasma levels increased. The basis for this disparity is unclear but may
reflect the differences in the ethiology of HF in these
two animal models. It is also possible that in the present
model, normal levels of PRA could indicate a state of
adequate compensation.
This model may be well suited for studying the pathophysiology of heart failure mediated by loss of contractile myocardium and for the evaluation of the efficacy of
pharmacological and other therapeutic interventions.
For instance, this model has been used to examine the
effects of long-term monotherapy with enalapril, metoprolol and digoxin on the progression of LV systolic dysfunction and LV chamber enlargement.36 This study
proved that in dogs with reduced LV ejection fraction,
early long-term monotherapy with enalapril or metoprolol prevents the progression of LV systolic dysfunction
and arrests, or attenuates, the process of progressive
chamber enlargement. In contrast, early therapy with
digoxin, while preventing the progressive decline in LV
Experimental Models of Heart Failure
Arrhythmia development in dogs at high risk
for sudden death with post-MI LV dysfunction
VTs /
3 weeks
6 weeks
8 weeks
Fig. 2 Diagram describing the arrhythmia progression in high risk dogs
after post-MI LV dysfunction has been produced by micrembolization.
Within the following eight weeks all dogs had suffered malignant
arrhythmia and sudden death. PVC, premature ventricular contraction;
SCD, sudden cardiac death; VTns, non-sustained ventricular tachycardia
(>5 and <20 consecutive beats); VTs, sustained ventricular tachycardia
(From Ref. 43).
sessed in dogs after MI and proved to be lower in susceptible dogs compared with the resistant group as already
Afterwards, ischemic left ventricular dysfunction was
induced in the same dogs by repetitive microembolization of the circumflex coronary artery until LV ejection
fraction reached 35%. Once stable LV dysfunction was
obtained, the incidence of SCD was significantly higher
in susceptible dogs. This group developed premature
ventricular contractions (PVCs) within days of reaching
an LVEF of 35% and progressed rapidly to nonsustained
then sustained ventricular tachicardia. By eight weeks,
all suscepitble dogs died suddenly after having sustained
or nonsustained ventricular tachycardia before their
death (Fig. 2). Resistant dogs developed only PVCs over
a much longer period of time and only one dog died suddenly three weeks after reaching the target LV ejection
A first major difference that distinguished the two
groups was that arrhythmias progression in susceptible
dogs was associated with a persistent sinus tachycardia,
thus reflecting a significant chronic elevated sympathetic
drive. On the other hand, resistant dogs, despite an identical cardiac damage, did not change heart rate throughout the six month follow-up. Susceptible dogs had
significantly lower vagal reflex activation measured by
baroreflex sensitivity (9.71.5 ms/mmHg susceptible vs.
28 ± 9.8 ms/mmHg, p < 0.01). Furthermore, susceptible
dogs had a marked sympathetic activation in response
to acute myocardial ischemia as indicated by the fact
that heart rates went from 220 ± 19 bpm at coronary
occlusion to 265 ± 18 bpm at 30 s of ischemia
(p < 0.05). In contrast, resistant animals had controlled
heart rates during exercise and coronary occlusion (from
218 ± 14 bpm at coronary occlusion to 231 ± 19 bpm at
30 s of ischemia, p = ns).
Another aspect that differentiated the two groups
concerned the ventricular repolarization. QT intervals
from susceptible dogs were longer after MI and prolonged
within eight weeks after LV dysfunction was established
(from 246 ± 26 to 274 ± 56 ms, p < 0.01). In contrast,
QT intervals in resistant dogs prolonged to a lesser degree only after 24 ± 6 weeks (from 231 ± 20 to 247 ± 20
ms, p = 0.03).
This first study described a new chronic animal model
of ischemic LV dysfunction in which dogs at high risk for
spontaneous SCD can be reliably identified. The main
findings are that autonomic imbalances (depressed vagal
and elevated sympathetic control of heart rate) present
before HF develops in subjects with ischemic heart disease, are associated with lethal arrhythmias as LV dysfunction progresses. Moreover abnormal repolarization
(QT interval prolongation and the loss of repolarization
adaptation to short cycle lengths) complete the high-risk
matrix. The implication of this study is that the autonomic and electrophysiological conditions resulting from
an MI, even in cases in which the degree of LV damage is
limited as in the present model, are critical in determining the outcome once LV dysfunction occurs. It seems
that an elevated sympathetic reflex response to ischemia determines a marked hypertrophic response and
Downloaded from by guest on December 3, 2014
around the circumflex branch of the left coronary artery.
After thirty days recovery from the MI the dogs are studied and characterized for developing ventricular fibrillation during an exercise and myocardial ischemia test on a
motor-drive treadmill. Each dog is exercised submaximally for 12–15 min while the work load increases progressively every 3 min until heart rate reaches a target
range of 215–225 beats/min. At that time, left circumflex artery is occluded for 2 min; the treadmill is stopped
after the first min of occlusion, while ischemia is maintained for an additional minute. The occurrence or not
of ventricular fibrillation (Fig. 1) during the two min of
exercise and myocardial ischemia clearly defines two
groups of dogs at high risk (SUSCEPTIBLE) and at low risk
for sudden cardiac death (RESISTANT). The availability of
two groups of dogs with opposite arrhythmia risk profile
makes this model unique and allowed the understanding
of critical autonomic mechanisms in sudden death. This
model indeed generated the first experimental evidence
that the analysis of reflex control of heart rate could provide meaningful prognostic information for risk stratification of post-MI individuals.39,40 This finding has been
successfully applied clinically41 and, now a days, baroreflex sensitivity and heart rate variability have a class I A
indication for risk stratification of sudden death.42 The
unique feature of this chronic animal model has been recently exploited into the major issue of chronic LV dysfunction and HF adding a chronic ischemic damage by
multiple microembolizations.43 Sudden cardiac death
risk was assessed in 15 dogs with a healed anterior MI
by the sub-maximal exercise and brief acute circumflex
ischemia test: six dogs were susceptible to SCD (i.e. they
developed VF) and were successfully defibrillated and
nine were resistant. Baroreflex sensitivity (BRS) was as-
a significant delay in the hypertrophied basal anterior
wall (239 ± 42 ms), which was longer than other areas
of the susceptible ventricle (scar 212 ± 26 ms, p < 0.01,
septum 222 ± 28 ms, p < 0.05 and lateral wall 197 ± 35,
p < 0.001, Fig. 3). The basal anterior wall of resistant
dogs did not have echocardiographic or electrophysiologic (electrograms with voltage >10 mV) evidence of
hypertrophy and repolarization was shorter (200 ± 21
ms, p < 0.001) when compared to susceptible dogs. In
resistant dogs, repolarization was also shorter in the septum (203 ± 21 ms resistant vs. 222 ± 28 ms susceptible,
p < 0.05). No significant regional repolarization differences were found in resistant dogs (Fig. 3). This study
demonstrated that regional heterogeneity of ventricular
repolarization after MI is an important component of the
matrix conducive to development of ischemia dependent
lethal arrhythmias. The current data set provides a direct
link between post-MI adverse remodeling, enhanced by
augmented sympathetic activation, and the electrical
consequences on ventricular repolarization associated
with high sudden death risk. QT intervals on the surface
electrocardiogram were prolonged in high-risk animals as
previously demonstrated,45 and are prolonged in humans
at high SCD risk.48,49 However, the extent of repolarization heterogeneity was not evident by the mild QT interval prolongation observed.
Based on findings from this study it is possible that
surface ECG analysis methods may underestimate the degree of ventricular repolarization abnormalities that contribute to arrhythmogenesis, thus decreasing the
measurements’ predictive value.50 Furthermore, the
data underscore the need for an accurate understanding
of repolarization alterations in the ischemic heart in light
of the extensive use of antiarrhythmic and non-cardiovascular drugs that act on repolarizing currents.
The spontaneous ventricular arrhythmia observed in
this experimental preparation is most consistent with a
reentrant mechanism. Reentrant arrhythmias require
unidirectional block and a path of slowed conduction.51
Regional differences in repolarization found in susceptible dogs provide the potential for unidirectional block
and the increased refractory gradient in susceptible dogs
theoretically increased the likelihood of unidirectional
block leading to reentry.
Post-MI depression of BRS and the tachycardic response
to acute ischemia documented that susceptible dogs respond to perturbations with sympathetic activation,
whereas resistant dogs had stronger vagal reflexes that
controlled heart rate during the same acute stimuli following the similar coronary ligation. These observations,
obtained in dogs with comparable LV dysfunction, suggest
that individual differences in post-MI neural activation are
critical to development of a myocardial substrate conducive for lethal arrhythmias. Adverse remodeling after MI is
well known, and it is traditionally described in terms of
histological52–54 or architectural changes55 over time.
Only a few data sets focus on the possibility that the electrical properties of the ventricle remodel after MI or how
this remodeling leads to sudden death.56,57 Integrating all
the changes resulting from the stimulus of myocardial injury may help elucidate the mechanisms responsible for
Downloaded from by guest on December 3, 2014
electrophysiological derangements that ultimately create the condition for an elevated arrhythmic risk.
The important issue of abnormal ventricular repolarization has been addressed in a very recent study in this
same model. Endocardial repolarization has been analysed in detail by the use of electrical mapping.44
In a group of 12 dogs surface QT intervals were obtained at similar cycle lengths to avoid the need of correction algorithms 30 days after MI. The mean RR
intervals used for the measurements were 421 ± 21 ms
for susceptible and 423 ± 21 ms for resistant dogs. Surface QT intervals were longer in susceptible dogs
(240 ± 10 ms) compared to resistant animals (222 ± 7
ms, p = 0.04).
Electrophysiological studies were performed through
a femoral arteriotomy during anesthesia (propofol 10
mg bolus induction followed by 2–5 mg boluses for
maintenance) using sterile technique. Descriptions of
the electroanatomic mapping system (CARTO, Biosense
Webster) are.46,47 With the use of a magnetic sensor,
the system displayed the location of the mapping catheter relative to the location of a reference sensor taped
to the posterior chest (spatial accuracy 61.0 mm). One
hundred to 300 local electrograms were obtained by first
confirming adequate catheter contact. Local repolarization time was defined as the interval between the R
wave peak and the end of the local T wave. This method
was used to decrease the chance for error in identifying
local activation. Each electrogram was visually examined and repolarization time was calculated by examining the digitally acquired signal using cursors to measure
the time from R wave peak to the end of the local T
wave. Regional repolarization kinetics were reconstructed in a 3-D color-coded map (longest-purple,
shortest-red). Two operators, blinded to the dog status,
determined local repolarization from the local
Local repolarization times were recorded from four
regions within the LV, which were chosen based on the
characteristics of regional remodeling following the
anterior MI: (1) the basal anterior wall, which was not directly involved with LAD ligation or acute circumflex ischemia; (2) the anteroapical area, which was chosen to
examine repolarization characteristics of the chronic infarct; (3) the lateral wall was evaluated as the area previously exposed to acute ischemia; (4) the anteroseptum,
which was chosen as a transition area between the apical
infarct and the basal anterior wall.
Histologically unique areas were also evaluated using
bipolar upstroke voltage as an indirect marker. Areas
with voltages <1.5 mV were considered scar tissue and
>10.0 mV were considered as hypertrophied areas. Repolarization duration was examined using five measurements selected from each histologic region.
Sinus cycle lengths during endocardial mapping were
similar in resistant (400 ± 44 ms) and susceptible
(397 ± 31 ms, p = 0.8) dogs. Average endocardial repolarization times using all regions were significantly longer in
susceptible (217 ± 36 ms) compared to resistant
(196 ± 21 ms, p < 0.001) dogs. Prolonged ventricular
repolarization in susceptible dogs was accounted for by
E. Vanoli et al.
Experimental Models of Heart Failure
Downloaded from by guest on December 3, 2014
Fig. 3 Endocardial repolarization duration maps from a resistant dog (left) and a susceptible animal (right) showing an anteroposterior projection (AP,
top), right anterior oblique (RAO, middle) and left anterior oblique (LAO, bottom). Local repolarization durations are color coded with dark-blue and
purple representing the longest duration (up to 200 ms) and orange representing the shortest repolarization duration (140 ms). Endocardial
repolarization duration was longest in the basal anterior wall of susceptible dogs (From ref. 44). (For interpretation of the references to color in this
figure legend, the reader is referred to the web version of this article.)
1. Sytkowski PA, Kannel WB, D’Agosrtino RB. Changes in risk factors and
declime in mortality from cardiovascular diseases: The Framingham
Heart Study. New Eng J Med 1990;322:1635–41.
2. Ghali JK, Cooper R, Ford E. Trends in hospitalization rates for heart
failure in the United States, 1973–1986. Evidence for increasing
population prevalence. Arch Int Med 1990;150:769–73.
3. National Center for Health Statistics. Detailed diagnosis and procedures, National Hospital Discharge Survey, 1990: Vital and Health
Statistics, Series 13, No. 113. Hyattsville, Md: The Center; 1992,
DHHS publication No. 92–1774.
4. Bourassa MG, Gurne
` O, Bangdiwala SI et al. Natural history and
patterns of current practice in heart failure. J Am Coll Cardiol
1993;22(Suppl A):14A–9A.
5. Packer M. The neurohormonal hypothesis: a theory to explain the
mechanism of disease progression in heart failure. J Am Coll Cardiol
6. Bouzamondo A, Hulot JS, Sanchez P et al. Beta-blocker benefit
according to severity of heart failure. Eur J Heart Fail 2003;5:281–9.
7. Gheorghiade M, Colucci WS, Swedberg K. Beta-blockers in chronic
heart failure. Circulation 2003;107:1570–5.
8. Hunt SA, Baker DW, Chin MH et al. ACC/AHA guidelines for the
evaluation and management of chronic heart failure in the adult:
executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines
(Committee to revise the 1995 guidelines for the evaluation and
management of heart failure). J Am Coll Cardiol 2001;38:
Khattar RS. Effects of ACE-inhibitors and b-blockers on left ventricular remodelling in chronic heart failure. Minerva Cardioangiol
Krum H. Tumor necrosis factor-a blockade as a therapeutic strategy
in heart failure (RENEWAL and ATTACH): unsuccessful, to be specific.
J Card Fail 2002;8:365–8.
Hongo M, Ryoke T, Ross Jr J. Animal models of heart failure: recent
Ellis Jr PR, Bailas NJ, Viskos JD et al. Experimental heart failure in
dogs. Arch Surg 1964;89:299–306.
Arnolda L, McGrath B, Cocks M et al. Adriamycin cardiomyopathy in
the rabbit: an animal model of low output cardiac failure with
activation of vasoconstrictor mechanisms. Cardiovasc Res 1985;19:
Franciosa JA, Heckel R, Limas C et al. Progressive myocardial
dysfunction associated with increased vascular resistance. Am J
Physiol 1980;239(Heart Circ Physiol 8):H477–82.
Armstrong PW, Stopps TP, Ford SE et al. Rapid ventricular pacing in
the dog: pathophysiologic studies of heart failure. Circulation
Flaim SF, Minteer WJ, Nellis SH et al. Chronic arteriovenous shunt:
evaluation of a model for heart failure in rat. Am J Physiol
1979;236(Heart Circ Physiol 5):H698–704.
Gaasch WH, Zile MR, Hoshino PK et al. Stress-shortening relations and
myocardial blood flow in compensated and failing canine hearts with
pressure overload hypertrophy. Circulation 1989;79:872–83.
Starzl TE, Gaertner RA. Chronic heart block in dogs. A method for
producing experimental heart failure. Circulation 1955;12:
Carlyle PF, Cohn JN. A nonsurgical canine model of chronic left
ventricular myocardial dysfunction. Am J Physiol 1983;244(Heart
Circul Physiol 13):H769–74.
Whipple GH, Sheffield LT, Woodman EG et al. Reversible congestive
heart failure due to chronic rapid stimulation of the normal heart.
Proc N Eng Cardiovasc Soc 1962;20:39–40.
Coleman HN, Taylor RR, Pool PE et al. Congestive heart failure
following chronic tachycardia. Am Heart J 1971;81:790–8.
Riegger AJ, Liebau G. The renin–angiotensin–aldosterone system.
Antidiuretic hormone and sympathetic nerve activity in an experimental model of congestive heart failure in the dog. Clin Sci
Wilson JR, Douglas P, Hickey WF et al. Experimental congestive
heart failure produced by rapid ventricular pacing in the dog: cardiac
effects. Circulation 1987;75:857–67.
Damiano RJ, Tripp HF, Asano T et al. Left ventricular dysfunction
and dilatation resulting from chronic supraventricular tachycardia.
J Thorac Cardiovasc Surg 1987;94:135–43.
Howard R, Stopps TP, Moe GW et al. Recovery from heart failure:
structural and functional analysis in a canine model. Can J Physiol
Pharmacol 1988;66:1505–12.
Spinale FG, Hendrick DA, Crawford FA et al. Chronic supraventricular tachycardia causes ventricular dysfunction and subendocardial
injury in swine. Am J Physiol 1990;259:H218–29.
Chow E, Woodard JC, Farrar DJ. Rapid ventricular pacing in pigs: an
experimental model of congestive heart failure. Am J Physiol
Shinbane JS, Wood MA, Jensen DN et al. Tachycardia-induced
cardiomyopathy: a review of animal models and clinical studies.
J Am Coll Cardiol 1997;29:709–15.
Pak PH, Nuss HB, Tunin RS et al. Repolarization abnormalities,
arrhythmia and sudden death in canine taachycardia-induced cardiomyopathy. J Am Coll Cardiol 1997;30:576–84.
Kaab S, Nuss HB, Chiamvimonvat N et al. Ionic mechanism of action
potential prolongation in ventricular myocites from dogs with pacinginduced heart failure. Circ Res 1996;78:262–73.
Feola M, Haiderer O, Kennedy JH. Experimental graded pump failure
of the left ventricle. J Surg Res 1971;11:325–41.
Lloyd-Jones W, Smith Jr SC, Urschel CW et al. Chronic heart failure
secondary to akinesis of the left ventricle: an experimental study.
Surgery 1973;74:607–16.
Downloaded from by guest on December 3, 2014
clinical observations that beta-blocker and ACE inhibitor
therapy after MI not only reduce adverse remodeling of
the ventricle and progression of ischemic heart disease,
but also reduce sudden death.
In the present experimental preparation differences
in repolarization heterogeneity between susceptible
and resistant dogs may be accounted for by two mechanisms: (1) different tissue responses to autonomic
changes associated with the MI. Indeed, autonomic markers (heart rate, baroreflex sensitivity and heart rate variability) in susceptible animals are different compared to
resistant dogs even before MI39,40 and respond differently
to both acute ischemia38 and chronic infarction over
time;58 (2) development of myocardial hypertrophy. This
is important since downregulation of repolarizing K+ currents is associated with left ventricular hypertrophy59
resulting in heterogeneity of repolarization currents.
Simulation models predict that action potential duration
would be prolonged with heterogeneity of Ikr and Iks
expression.60 It is conceivable that the differences in
neural activation seen in this study are key mechanisms
in the electrical and histological remodeling that led to
high VF risk. In this model, interruption of local sympathetic input to the heart by left stellectomy reduces
the incidence of VF61 suggesting that imbalanced sympathetic input to the ventricles is an important factor in
arrhythmogenesis.62 It is possible to speculate, therefore, that increased sympathetic activity in susceptible
dogs resulted in regional hypertrophy and downregulation of repolarizing potassium channels in specific areas,
which lead to heterogeneity of repolarization. These
findings represent a bridge between the large body of
data primarily focusing on tissue and structural remodeling following MI and how the remodeling itself creates
risk for lethal arrhythmias in the context of autonomic
imbalance due to a previous MI.
E. Vanoli et al.
Experimental Models of Heart Failure
47. Schwartz PJ, Wolf S. QT interval prolongation as predictor of sudden
cardiac death in patients with myocardial infarction. Circulation
48. Schouten EG, Dekker JM, Meppelink P et al. QT interval prolongation
predicts cardiovascular mortality in an apparently healthy population. Circulation 1991;84:1516–23.
49. Algra A, Tijssen JG, Roelandt JR et al. QTc prolongation measured
by standard 12-lead electrocardiography is an independent risk
factor for sudden death due to cardiac arrest. Circulation
50. Malik M, Batchvarov VN. Measurement, interpretation and clinical
potential of QT dispersion. J Am Coll Cardiol 2000;36:1749–66.
51. The Sicilian Gambit. A new approach to the classification of
antiarrhythmic drug based on their actions on arrhythmogenic
mechanism. Task force of the working group on arrhythmias of the
European Society of Cardiology. Circulation 1991;84:1831–51.
52. Pfeffer MA, Braunwald E. Ventricular remodelling after myocardial
infarction: experimental observations and clinical implication.
Circulation 1990;81:1161–72.
53. Kramer CM, Rogers WJ, Park CS et al. Regional myocite hypertrophy
parallels regional myocardial dysfunction during post-infarction
remodelling. J Mol Cell Cardiol 1998;30:1773–8.
54. Mitchell GF, Lamas GA, Vaugham DE et al. Left ventricular remodelling in the year after first anterior myocardial infarction. J Am Coll
Cardiol 1992;19:1136–44.
55. Eaton LW, Weiss JL, Bulkley BH et al. Regional cardiac dilatation
after acute myocardial infarction. N Eng J med 1979;300:57–62.
56. Cao J, Fishbein MC, Han JB et al. Relationship between regional
cardiac hyperinnervation and ventricular arrhythmia. Circulation
57. Gaudron P, Kugler I, Hu K et al. Time course of cardiac
structural, functional and electrical changes in asymptomatic
patients after myocardial infarction: their inter-relation and
prognostic impact. J Am Coll Cardiol 2001;38:33–40.
58. Adamson PB, Huang MH, Vanoli E et al. Unexpected interaction
between beta-adrenergic blockade and heart rate variability before
and after myocardial infarction; A longitudinal study in dogs at high
and low risk for sudden death. Circulation 1994;90:976–82.
59. Volders PG, Sipido KR, Vos MA et al. Downregulation of delayed
rectifier K(+) currents in dogs with chronic complete atrioventricular
block and acquired torsades de pointes. Circulation 1990;100:
60. Viswanathan PC, Shaw RM, Rudy Y. Effects of Ikr and Iks heterogeneity
on action potential duration and its rate dependence: a simulation
study. Circulation 1999;99:2466–74.
61. Schwartz PJ, Stone HL. Left stellectomy in the prevention of ventricular fibrillation due to acute myocardial ischemia in conscious dogs with
an anterior myocardial infarction. Circulation 1980;100:2455–61.
62. Schwartz PJ. QT prolongation, sudden death and sympathetic
imbalance: the pendulum swings. J Cardiovasc Electrophysiol
Downloaded from by guest on December 3, 2014
33. Polansky J, Mandell G, Wasfie T et al. Prolonged ventricular
dysfunction induced by sequential multifocal ischemic infarcts in
dogs. Artif Organs 1984;8:445–57.
34. Weber KT, Malinin TI, Dennison BH et al. Experimental myocardial
ischemia and infarction: production of diffuse myocardial lesions in
unanesthetized calves. Am J Cardiol 1972;29:793–802.
35. Sabbah HN, Stein PD, Kono T et al. A canine model of chronic heart
failure produced by multiple sequential coronary microembolizations. Am J Physiol 1991;260(Heart Circ Physiol 29):H1379–84.
36. Sabbah HN, Shimoyama H, Kono T et al. Effect of long-term
monotherapy with enalapril, metoprolol and digoxin on the progression of left ventricular dysfunction and dilatation in dogs with
reduced ejection fraction. Circulation 1994;89:2852–9.
37. Sabbah HN, Goldberg AD, Schoels W et al. Spontaneous and
inducible ventricular arrhytmias in a canine model of chronic
heart failure: relation to haemodynamics and sympathoadrenergic
activation. Eur Heart J 1992;13:1562–72.
38. Schwartz PJ, Billman GE, Stone HL. Autonomic mechanisms in
ventricular fibrillation induced by myocardial ischemia during
exercise in dogs with healed myocardial infarction. An experimental preparation for sudden cardiac death. Circulation 1984;69:
39. Schwartz PJ, Vanoli E, Stramba-Badiale M et al. Autonomic mechanisms and sudden death. New insights from analysis of baroreceptor
reflexes in consciuos dogs with and without a myocardial infarction.
Circulation 1998;78:969–79.
40. Hull Jr SS, Evans E, Vanoli E et al. Heart rate variability before and
after myocardial infarction in dogs at high and low risk for sudden
cardiac death. J Am Coll Cardiol 1990;16:978–85.
41. La Rovere MT, Bigger JT, Marcus F et alfor the ATRAMI (Autonomic
Tone and Reflexes After Myocardial Infarction) Investigators. Baroreflex sensitivity and heart rate variability in prediction of total
42. Priori SG, Aliot E, Blomstrom-Lundqvist C et al. Task force on sudden
cardiac death of the European Society of Cardiology. Eur Heart J
43. Adamson PB, Vanoli E. Early autonomic and repolarization abnormalities contribute to lethal arrhythmias in Chronic Ischemic heart
failure. Characteristics of a novel heart failure model in dogs with
postmyocardial infarction left ventricular dysfunction. J Am Coll
Cardiol 2001;37:1741–8.
44. Swann MH, Nakagawa H, Vanoli E et al. Heterogeneus regional
endocardial repolarization identifies increased risk for ischemia
dependent ventricular fibrillation after myocardial infarction. J
Cardiovasc Electr 2003;14:873–9.
45. Ben-Haim SA, Osadchy D, Schuster I et al. Nonfluoroscopic, in vivo
navigation and mapping technology. Nat Med 1996;2:1393–5.
46. Gepstein L, Hayam G, Ben-Haim SA. A novel method for nonfluoroscopic catheter-based electroanatomical mapping of the heart: in
vitro and in vivo accuracy results. Circulation 1997;95:1611–22.