Left Ventricular Performance and Remodeling in Rabbits After Myocardial Infarction
Effects of a Thyroid Hormone Analogue
Background Because the rat postinfarction model differs from human heart failure with respect to the composition of myosin heavy chain (MHC) isoforms and other contractile proteins, alternative animal models are needed for the development of new treatments for human heart failure. The purpose of this study was threefold: (1) to test the feasibility of using the V3(β,β) rabbit postinfarction model for the study of heart failure by characterizing the effects of chronic coronary artery occlusion on the left ventricle; (2) to determine whether the thyroid hormone analogue 3,5-diiodothyropropionic acid (DITPA) produces improvements in left ventricular function; and (3) to determine the effects of myocardial infarction and treatment with DITPA on MHC protein isoforms.
Methods and Results Male New Zealand White rabbits underwent proximal circumflex coronary artery ligation. After infarction, rabbits were treated with DITPA (3.75 mg/kg body wt) or placebo for 21 days and then underwent conscious and open-chest hemodynamic studies. In separate groups of rabbits, β- and α-MHC isoforms were separated, and relative proportions were measured using gradient sodium dodecyl sulfate–polyacrylamide gel electrophoresis and laser densitometry. Infarction resulted in increased left ventricular end-diastolic pressure and prolonged left ventricular relaxation (τ) (P=.001 for both variables). Postinfarction treatment with DITPA decreased left ventricular end-diastolic pressure and τ (P=.002 and P=.001, respectively) and increased maximum positive and negative dP/dt (P=.002 and P=.016, respectively). Infarcted rabbits treated with DITPA had no significant changes in heart rate or left ventricular systolic pressure compared with untreated rabbits with infarction. There were no significant differences in heart rate, positive dP/dt, peak systolic pressure, or τ between sham-operated rabbits and sham-operated rabbits treated with DITPA. Although infarction resulted in increased left ventricular diameter, there were no effects of DITPA on left ventricular remodeling. Neither myocardial infarction nor treatment with DITPA altered the ratio of MHC isoforms.
Conclusions Rabbits that survive occlusion of the circumflex artery will develop myocardial dysfunction and left ventricular remodeling. Therapy with DITPA, a thyroid hormone analogue, produces improvement in ventricular performance and reduces end-diastolic pressure. The hemodynamic effects of DITPA were not associated with alterations of MHC isoforms. Whether DITPA represents the prototype of a previously undescribed class of agents for the treatment of heart failure will need to be determined by clinical trials.
The rat model of coronary artery ligation is the most widely studied model of chronic left ventricular infarction. With respect to alterations in left ventricular function,1 2 remodeling,2 3 4 neurohumoral and molecular effects,5 6 and long-term survival,7 the rat postinfarction model shares many pathophysiological characteristics with human ischemic heart failure.8 9 10 Unlike human heart failure, however, left ventricular dysfunction in the rat model is associated with alterations in myosin heavy chain (MHC) isoform composition.11 During chronic adaptation to excess hemodynamic loading, MHC is partially downregulated in the rat from a predominantly V1(α,α) to a predominantly V3(β,β) isoform.
MHC composition is upregulated from the V3 to the V1 isoform by thyroid hormone, which binds to cardiac nuclear receptors and alters gene expression via transcriptional regulation.12 13 Earlier, we have shown that when thyroid hormone was administered to rats with heart failure after myocardial infarction, left ventricular performance was improved, an effect that was associated with upregulation of the V3 to the V1 isoform.11 Recently, we showed that 3,5-diiodothyropropionic acid (DITPA), an analogue of thyroid hormone, binds to the bacterially expressed thyroid hormone nuclear receptor subtypes α1 and β1 and induces α-MHC mRNA expression in cultured cardiac myocytes.14 In hypothyroid rats, it produces positive inotropic activity with minimal effects on heart rate and metabolic activity.14 More recently, we demonstrated that DITPA also improved cardiac performance when given, in conjunction with captopril, to rats with heart failure.15
Because rats differ from humans with respect to MHC isoenzyme composition, there also may be other important differences in proteins necessary for excitation-contraction coupling. Other investigators have shown that thyroid hormone alters other contractile proteins in addition to MHC.16 Using MHC composition as a biochemical marker, we hypothesized that an animal model with a V3(β,β) predominance would provide a more relevant model for studying human heart failure.
No model of cardiac dysfunction or remodeling after chronic coronary artery occlusion has been characterized in a predominantly V3 animal. We have recently developed a postinfarction model of left ventricular dysfunction in the rabbit, an animal in which V3 is the predominant MHC isoform. The purpose of this study was threefold: (1) to test the feasibility of using the V3 rabbit postinfarction model for the study of heart failure by characterizing the effects of chronic coronary artery occlusion on the left ventricle; (2) to determine whether monotherapy treatment with DITPA produces improvements in left ventricular function; and (3) to determine the effects of myocardial infarction and treatment with DITPA on MHC isoforms.
Male New Zealand White rabbits (3 to 4 kg) were used for all experiments. Animals were housed in separate cages in an environmentally controlled facility with appropriate light-dark cycles and were maintained on standard rabbit chow and given water ad libitum. All animals were treated in accordance with guidelines of the American Association for the Accreditation of Laboratory Animal Care.
The experimental protocol consisted of four groups of rabbits: sham (thoracotomy, circumflex artery not ligated); sham-DITPA (sham, treated with thyroid hormone analogue); myocardial infarction (MI, thoracotomy, circumflex artery ligated); and MI-DITPA (thoracotomy, circumflex artery ligated, treated with thyroid hormone analogue). In alternating order, rabbits were preassigned to each group on the day of surgery.
Circumflex Artery Ligation
The method of circumflex artery ligation was adapted from previously described techniques.17 18 All surgical procedures were performed in a sterile manner in the animal operating suites at our facility. The rabbits were anesthetized in a halothane inhalant chamber and an intramuscular injection of a ketamine (0.4 mg) and acepromazine (20 mg) mixture was administered. Animals were intubated and ventilated with a Harvard Rodent ventilator (model 683) and monitored by continuous ECG tracing using standard limb leads. Intravenous access was established.
The rabbits were placed on a warming blanket, and the chest was shaved and cleansed in sterile fashion. A left thoracotomy was performed through the fifth intercostal space, and the circumflex artery was exposed. A suture was placed around the artery at the origin of the vessel such that it could be loosened. For sham-operated animals, the suture was placed but removed at this time, and the chest was closed with layered sutures.
To prevent ventricular arrhythmias at the time of ligation of the circumflex artery, a series of repeated occlusions and reperfusions were done.17 First, a 2-minute occlusion was performed, and the extent of ischemia was visually and electrocardiographically assessed. After a 5-minute reperfusion period, the artery was again occluded, for 5 minutes, followed by 10 minutes of reperfusion. A third occlusion for 10 minutes was then performed, followed by 15 minutes of reperfusion. Finally, the artery was permanently ligated. Lidocaine (1 mg/kg IV) was administered during the final 15-minute reperfusion period or before closing the chest in sham-operated animals. If adequate ECG evidence for ischemia (ST-segment elevation of more than 3 mm in leads II, III, and aVF and/or I and aVL) was not appreciated, an additional ligature was applied to a branch of the circumflex or left anterior descending coronary artery using the same series of repeated occlusions and reperfusions. The chest was closed, and the rabbit was given a one-time subcutaneous dose of Bicillin (300 000 U) and was observed for several hours until stable.
In the event of sustained ventricular tachycardia or ventricular fibrillation during coronary occlusion or in the immediate postinfarction time period, the animals were given additional lidocaine (1 mg/kg IV). Cardioversion or defibrillation was performed as needed with 5 J of DC current using defibrillator paddles applied to the chest.
Fig 1⇓ shows a photomicrograph of a typical trichrome-stained thin section of the left ventricle from a sham-operated and infarcted rabbit (Fig 1A⇓ and 1B⇓, respectively). The section was taken one third of the distance from the midpoint of the ventricular anteroposterior axis toward the apex. The infarcted portion is seen to lie along the posterolateral and apical portion of the left ventricular wall.
Three weeks after surgery, conscious hemodynamic measurements were obtained. Rabbits were anesthetized with halothane in an inhalant chamber, and a 5F micromanometer-tipped catheter (Millar Instruments) was inserted into the left ventricle via the right carotid artery under constant pressure monitoring. The zero-pressure baseline was obtained by placing the pressure sensor in 37°C saline. The ventricular catheter and a central venous catheter (PE 90) were inserted through the right internal jugular vein and tunneled subcutaneously to the dorsum of the animal. After a 4-hour recovery period in a rabbit cage, heart rate, left ventricular systolic pressure, and left ventricular end-diastolic pressure were recorded on a physiological recorder (model 2400, Gould Instrument) and an IBM AT microcomputer in which the high-frequency cutoff filter was set at 100 Hz. Left ventricular pressure from 60 to 90 consecutive cardiac cycles was digitized to ensure reproducibility; averaged data are reported. Heart rate, rate of maximum positive and negative left ventricular pressure development (dP/dt), and isovolumic relaxation time (τ) were determined using customized software as previously reported from our laboratory.3 14 15
Open-Chest Determinations of Stressed Ventricular Performance
The rabbit was anesthetized with intravenous thiobutabarbital (500 mg), intubated, and ventilated using mechanical ventilation. A midsternotomy was performed. After left ventricular pressure and heart rate stabilized, the ascending aorta was exposed, and a snare was placed around the proximal vessel. While left ventricular systolic pressure was being recorded, abrupt occlusion was produced by sudden tightening of the snare. Peak developed pressure was defined as the mean systolic pressure minus the mean end-diastolic pressure over the first five stable beats after aortic occlusion.1 15
To determine baseline and maximum flow-generating capacity of the heart,1 2 warmed (39°C) Tyrode’s solution was infused (45 mL · min−1 · kg−1) while left ventricular end-diastolic pressure was recorded continuously. The infusion was continued for 90 to 120 seconds or until hemodynamic compromise was noted. Baseline open thorax end-diastolic pressure and postinfusion open-thorax end-diastolic pressure were recorded. Fifteen minutes after the volume infusion, when end-diastolic pressure had decreased to at least 75% of its postvolume peak, a precalibrated flow probe (3.0 mm id, Transonics Instruments) was placed around the ascending aorta, and the ventricular catheter was withdrawn into the right carotid artery so as not to interfere with flow-probe measurements. Baseline flow was then recorded (mL/min) as the integrated mean of pulsatile flow. Flow-generating capacity of the heart was then measured using warmed Tyrode’s solution, infused as described above. Maximum flow-generating capacity was defined as the plateau of mean flow, which was usually achieved after 70 to 90 seconds of infusion.
Measurements of Ventricular Remodeling
Pressure-volume data were recorded with methods described previously.1 2 3 4 Briefly, 1000 U heparin was administered intravenously followed by intravenous KCl (5 mL) to arrest the heart in diastole. The heart was rapidly removed and rinsed in 0.9% NaCl, and the right ventricle was incised. Two telescoped catheters (PE 10 inside/PE 90 outside for measurement of pressure and infusion of saline, respectively) were inserted in the left ventricle via the remnant aorta. The atrioventricular groove was identified, and a ligature was passed around the heart and tied to isolate the left atrium from the left ventricle. After gentle aspiration of the left ventricle to remove excess blood and to reduce the pressure to −5 mm Hg, normal saline was infused at a rate of 3.3 mL/min into the suspended left ventricle, and pressure was recorded continuously until the pressure increased to 40 mm Hg. The accumulated fluid was aspirated, and infusion was repeated. Three curves were obtained from each ventricle within 15 minutes of cardiac arrest.
The heart was removed and resuspended, and the coronaries were perfused with 10% buffered formalin via the infusion catheter after it was withdrawn into the aortic remnant. The ventricles were separated from the atria, and the right ventricular free wall was dissected free from the septum. The ventricles were weighed individually and stored in 10% buffered formalin.
The length of the left ventricle was determined by inserting a probe through the aortic root and passing it to the apex. The probe was marked at the level of the aortic valve, removed, and measured using an electronic hand-held micrometer (10−2 mm sensitivity). The left ventricle was then transected at the midpoint. The major and minor diameters of the basilar portion were measured using the micrometer. The thickness of the noninfarcted left ventricular anterior wall was measured by averaging the micrometer-measured mural thickness at two sites approximately 30 degrees apart.
An additional group of rabbits underwent coronary ligation or sham operation after randomization to treatment with DITPA or placebo injection. Three weeks after surgery, the animals were anesthetized, and left ventricular hemodynamics were measured according to the previously described procedure. The animals were then killed with an overdose of methoxyflurane, and the heart was removed. Left ventricles were separated and weighed. The left ventricles were immediately frozen in liquid nitrogen and placed in storage at −80°C until MHC composition was analyzed.
Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) was performed as described by Esser et al.20 In brief, left ventricular tissue was homogenized (1:10 wt/vol) in 62.5 mmol/L Tris (pH 6.8). Protein concentrations were determined from an aliquot of homogenate using the Peterson modification of the micro-Lowry assay.19 Rabbit atrial and soleus tissue were prepared in the same fashion for α- and β-MHC controls, respectively. Vertical, discontinuous SDS-PAGE was performed using a 4% to 9% linear gradient resolving gel and a 3% stacking gel (total acrylamide concentrations, respectively). Samples containing homogenate plus twofold concentration Laemmli buffer (15 μL total volume, 1.25 to 10 μg total protein) were electrophoresed with constant DC current starting at 25 mA/gel for 30 minutes and then increasing to 50 mA/gel for an additional 3 hours.
After electrophoresis, gels were placed in 10% (wt/vol) trichloroacetic acid for 15 minutes to limit protein diffusion and then stained with Coomassie blue overnight. The Coomassie blue solution consisted of equal volumes of solution A (0.3% [wt/vol] Coomassie blue R-250, 90% [vol/vol] ethanol) and solution B (1% [wt/vol] copper sulfate in 20% [vol/vol] glacial acetic acid). Gels were destained for 3 to 5 hours in a solution of 0.5% (wt/vol) copper sulfate, 10% (vol/vol) glacial acetic acid, and 25% (vol/vol) methanol. Gels were subsequently scanned on a laser densitometer (LKB UltraScan XL 2222-020). Relative areas under the α- and β-MHC peaks were calculated using system software that accompanied the densitometer.
According to previously described methods,1 2 3 4 the measured pressure (P)-volume (V) data were fitted to an exponential equation P=P0eKV, where K, the chamber stiffness constant, was derived from a plot of lnP=KV+lnP0. Left ventricular end-diastolic volume index was defined as the infused ex vivo left ventricular volume measured at pressure equal to in vivo end-diastolic pressure. Left ventricular chamber volume (V) was defined as ventricular cavity volume at a distending pressure of 10 mm Hg, as determined from the pressure-volume relation. Left ventricular wall volume, VW, was determined from the mass of the left ventricle, such that VW is left ventricular mass (g)/1.06 (density of muscle). The ratio of V to VW was then calculated as a measure of relative ventricular dilatation and hypertrophy.
After the morphometric measurements were taken, the left ventricle was cut from apex to base in three transverse slices and embedded in paraffin. Thin sections of the left ventricle were stained with Masson’s trichrome. The three sections were projected using an overhead projector and traced on 8×11-in sheets of paper. The circumferences of the infarcted and noninfarcted segments of the epicardial and endocardial surfaces were measured by digitization using an IBM AT compatible personal computer equipped with customized software. Infarct size is reported as the mean percent of the endocardial and epicardial circumferences involved with scar tissue for the three sections.1 2 3 4
For experiments, stock solutions of DITPA (Sigma Chemical Co) were prepared by dissolving the powder in concentrated sodium hydroxide and diluting to 20 mg/mL with distilled water.14 15 Hydrochloric acid was used to adjust the pH to 7 to 8. Animals were treated with a daily subcutaneous injection of DITPA, 3.75 mg/kg body wt, or an equivalent volume of 0.9% saline for 21 days. Treatment was started on the first postoperative day and continued to the morning of study.
Values are given as mean±SD. To evaluate whether there was an interaction between DITPA treatment and infarction, all data were analyzed by a two-factor ANOVA. The analysis tested the main effects (DITPA, yes/no; infarction, yes/no) and an interaction between DITPA and infarction. We also analyzed sham versus sham-DITPA, and infarction versus infarction-DITPA, using separate unpaired t tests. Significance was defined at the P<.05 level.
A total of 101 rabbits underwent initial sham (n=45) or coronary artery ligation (n=56) surgery. There were 6 immediate deaths in sham-operated animals, due to either perioperative or postoperative bleeding complications or complications due to general anesthesia. Most of these deaths occurred in the early stages of the study. Seven additional rabbits died during the 3-week postoperative period; 3 deaths were spontaneous and 4 deaths occurred during instrumentation before measurement of conscious hemodynamics. Among rabbits undergoing coronary artery ligation, 15 rabbits died during surgery or during the immediate postoperative period. The majority of these deaths were due to refractory ventricular fibrillation or cardiogenic shock (33% surgical mortality). Four additional untreated rabbits died in their cages during the 3-week postoperative period, and 1 DITPA-treated rabbit died during this period. A total of 3 additional infarcted rabbits, 2 untreated and 1 treated, died during instrumentation for conscious hemodynamic studies.
There were no significant differences in final body weights (kg) among the four groups of rabbits: 3.89±0.43, 3.78±0.41, 3.76±0.16, and 3.64±0.19, for sham, infarct, sham-DITPA, and infarct-DITPA, respectively. Percent weight gain, from time of surgery to day of study, also was similar among the four groups: 4.6±6.4%, 5.8±7.1%, 5.8±4.3%, and 3.2±7.5%. Mean infarct size was similar between untreated and DITPA-treated rabbits (24±6% versus 24±7%; range, 15% to 33% and 17% to 35%, respectively). Conscious left ventricular hemodynamics in the four groups of rabbits are shown in Table 1⇓. With respect to the main effects of infarction, left ventricular end-diastolic pressure and left ventricular relaxation (τ) were increased (P=.001 for both variables) after coronary artery ligation. The main effects of DITPA treatment were to decrease left ventricular end-diastolic pressure and τ (P=.002 and P=.001, respectively) and to increase maximum positive and negative dP/dt (P=.002 and P=.016, respectively). There were no differences in heart rate, positive dP/dt, peak systolic pressure, or τ between sham-operated rabbits and sham-operated rabbits treated with DITPA. Peak negative dP/dt, however, was significantly increased in the sham-treated rabbits. Infarcted rabbits treated with DITPA had no changes in heart rate or systolic pressure compared with untreated rabbits with infarction. However, compared with untreated rabbits with infarction, left ventricular end-diastolic pressure and relaxation time (τ) were reduced 54% and 29%, respectively, toward normal (P=.005 for both variables) by DITPA treatment. Compared with untreated infarcted rabbits, maximum positive and negative dP/dt were increased (P=.005 and P=.001, respectively) in DITPA-treated rabbits with infarction. There was a significant interaction between infarction and DITPA treatment for left ventricular end-diastolic pressure and τ, indicating that in the presence of infarction, DITPA treatment had specific effects on these variables.
Open-chest measurements of left ventricular performance are shown in Table 2⇓. With respect to the main effects of infarction, rabbits with infarction had decreased (P=.042) maximum flow generating capacity (cardiac index stress) and elevated end-diastolic pressure at peak flow (P=.002). Treatment with DITPA had no effects on any baseline or stress variable, although there was a trend for peak developed pressure and peak flow to be increased in treated rabbits. In sham-operated rabbits, there was no effect of DITPA treatment on peak developed pressure, cardiac index at rest or with stress, or baseline end-diastolic pressure or end-diastolic pressure at peak flow. When infarcted rabbits treated with DITPA were compared with untreated infarcted rabbits, there was an improvement in peak developed pressure and cardiac index stress in treated animals (P=.017 and P=.005, respectively). Data from resting and stress cardiac index, in the form of cardiac function curves, are displayed in Fig 2⇓. Treatment of infarcted rabbits with DITPA shifted the depressed curve in untreated controls upward and to the left. There were no significant interactions between infarction and treatment on any variable of stress cardiac function.
Derived indexes of left ventricular remodeling are shown in Table 3⇓. There was a trend toward increased left ventricular end-diastolic volume index (P=.051) after infarction. There were no effects, however, of either infarction or treatment on the chamber stiffness constant (Kc) or the ratio of left ventricular cavity volume (at pressure of 10 mm Hg) to left ventricular wall volume (V/Vw). There were no interactions between infarction and treatment on remodeling parameters, nor were there any significant differences between untreated animals and respective treated controls.
Measured indexes of cardiac remodeling are displayed in Table 4⇓. Left ventricular diameter was increased (P=.002) after infarction, whereas treated animals showed a trend toward increased left ventricular length. Neither infarction nor treatment produced any differences in heart weights or in left ventricular wall thickness. There were no differences between untreated animals and their respective treated controls in any measured index of left ventricular remodeling.
In rabbits assigned to the protocol for assay of MHC composition, unconscious left ventricular hemodynamics were obtained before they were killed and ventricular tissue was obtained. A total of 17 rabbits were studied: sham (n=4), sham-DITPA (n=4), infarct (n=5), and infarct-DITPA (n=4). There were no significant differences in heart rate between the four groups. Left ventricular systolic pressure and positive dP/dt were decreased and left ventricular end-diastolic pressure and τ were increased in the untreated infarct rabbits compared with sham, sham-DITPA, and MI-DITPA groups (P<.05 for all comparisons).
The percentage of α-MHC for the four groups was as follows: 39.5±1.3%, 43.3±1.5%, 41.4±7.5%, and 44.7±6.2% for sham, sham-DITPA, MI, and MI-DITPA, respectively. The percentage of β-MHC for the four groups was as follows: 60.5±1.3%, 56.7±1.5%, 58.6±7.5%, and 55.3±6.2% for sham, sham-DITPA, MI, and MI-DITPA, respectively. There were no significant differences in the proportion of either myosin isoform among the four groups.
Until recently, the coronary artery ligation model in rats has been the only suitable and widely available animal model in which to assess the long-term effects of myocardial infarction. Hof et al20 were the first to propose the rabbit coronary occlusion model as an alternative model and specifically assessed the effects of chronic infarction on circulating neurohormones and baroreflex responses. In the present study, we have provided data on the effects of myocardial infarction on conscious left ventricular hemodynamics and ventricular remodeling as well as on the pharmacological effects of the thyroid hormone analogue DITPA on ventricular performance and remodeling. We had previously shown that DITPA was effective when added to captopril in the rat postinfarction model of heart failure,15 and we sought to establish whether beneficial effects could be obtained with DITPA alone in the V3-dominant rabbit model of left ventricular dysfunction.
Long-term Effects of Myocardial Infarction in Rabbits
Using techniques similar to those reported earlier for measurement of infarct size,20 we confirmed that rabbits that survive for longer than 1 week after coronary occlusion generally have myocardial infarctions less than 30% in magnitude. In our study, there were only 4 rabbits with infarct sizes of more than 30%−two in the DITPA-treated group and two in the untreated group. The infarct size that is compatible with early survival and heart failure in rabbits is similar to the amount of damage reported in humans as causing compromised left ventricular function.21 By contrast, in the rat model, infarction sizes of more than 40% are routinely observed in animals that survive weeks or months1 2 3 4 7 until the time of experimental studies. The differences between infarct size in rat and rabbits undergoing coronary ligation are not clear but may be related to differences in coronary anatomy or postinfarction collateral development. The proximal level at which the artery is ligated, or the technique of ischemic preconditioning versus abrupt occlusion, may also be contributory.
Although we do not specifically report the relation of ST-segment shifts at the time of coronary occlusion to chronic infarct size, we observed that the amount of myocardial damage, as histologically assessed in the postinfarct period, bore little relation to either the extent of contiguous ECG abnormalities or the magnitude of acutely measured ST-segment elevation. Thus, unlike the rat postinfarction model, in which ECG screening can be highly specific for large infarction,1 2 3 4 7 alternative ECG or perhaps imaging techniques will need to be developed that can predict infarct size before death in the rabbit.
The present study also demonstrates the feasibility of obtaining extensive conscious hemodynamic measurements in rabbits with small or moderately large myocardial infarctions. The data from these hemodynamic studies show that rabbits with infarct sizes that average 26% develop marked elevations in left ventricular end-diastolic pressure, decreased rate of pressure development, and abnormal prolongation of isovolumic ventricular relaxation. Ventricular performance measured under conditions of maximum preload and afterload stress is also impaired, as manifested by a decrease in maximum left ventricular pressure development and flow generation. These results are similar to those seen in the rat infarct model.1 2
It is generally agreed that ventricular remodeling represents a cardiac response to chronic alterations in ventricular loading.22 In the present study, rabbits with myocardial infarction had significant increases in left ventricular midcavity diameter and a trend toward increases in left ventricular length and end-diastolic volume index. However, left ventricular wall thickness, measured in the noninfarcted anterior wall, was not decreased. The effect of these alterations in left ventricular geometry was to maintain a normal cavity volume–to–wall volume ratio (V/VW).
In summary, it is apparent that alterations in left ventricular systolic and diastolic function are present despite a modest but significant increase in left ventricular diameter and a trend toward increased end-diastolic volume. These data further suggest that although the size of myocardial infarction was large enough to promote cavity dilatation, compensatory ventricular hypertrophy occurred that was sufficient to maintain V/VW constant.
Effects of Thyroid Hormone Analogue DITPA
We earlier demonstrated in the rat postinfarction model that treatment with DITPA, when combined with captopril, increased cardiac index and −dP/dtmax and decreased left ventricular end-diastolic pressure and τ.15 In that study, DITPA was added to captopril because converting-enzyme inhibitors are the mainstay of treatment for heart failure. In the present study, which was designed to examine the effects of monotherapy, DITPA treatment alone produced significant improvement in resting and stress left ventricular hemodynamics in infarcted rabbits. The effects of infarction and treatment on resting and stress cardiac index have been summarized graphically in Fig 2⇑. It can be readily observed that the effects of DITPA are to shift the Frank-Starling relation of infarcted rabbits upward and to the left. These results are in contrast to those in infarcted rats treated with captopril alone, in which end-diastolic pressure and remodeling are decreased even though relaxation and global systolic indexes of resting and stress ventricular function are not improved.15
As in the rat model, the direct myocardial effects of DITPA are likely to be a major underlying reason for the improvement seen in cardiac performance in infarcted rabbits. First, heart rate was not significantly different between untreated and treated infarcted rabbits. Second, because left ventricular end-diastolic pressure was lower in treated rabbits and since both groups of infarcted rabbits had similar remodeling characteristics and volume stiffness, increased preload was probably not the reason for improvement in cardiac performance in DITPA-treated rabbits. Evidence supporting the contention that afterload was not decreased in DITPA-treated infarcted rabbits is that similar left ventricular systolic pressures at rest were observed between the two postinfarction groups, which was not associated with a change in resting cardiac index.
The effect of DITPA on left ventricular performance is most likely due to upregulation of thyroid hormone-responsive genes. However, the results of the gradient SDS-PAGE of cardiac MHCs suggest that the hemodynamic improvement of rabbits with left ventricular dysfunction after myocardial infarction does not depend on upregulation of β-MHC. Although this result was somewhat surprising, it was not wholly unexpected. First, the presence of left ventricular dysfunction in the untreated infarcted group did not appear to depend on further downregulation of V3, since percent β-MHC was not significantly altered in any group. Second, although we have shown that rats downregulate to V3 after large myocardial infarction and severe left ventricular dysfunction occurs,11 these results do not necessarily indicate a cause-and-effect relation. These results also should not be extrapolated to the rabbit model. Third, it is unclear whether improvement of left ventricular function in rats with ischemic heart failure treated with thyroxine11 or DITPA15 is directly related to upregulation of V3. For example, previous investigators have shown that preservation of normal myosin ATPase activity and myosin isoform distribution in infarcted rats was not associated with any apparent improvement in cardiac performance.23 An alternative explanation for the improvement in ischemic left ventricular dysfunction in both rats and rabbits after treatment with DITPA is the upregulation of other thyroid hormone–responsive genes, for example, the sarcoplasmic reticulum calcium ATPase and the ryanodine-sensitive release channel.16 24 25 Since it is known that calcium handling is altered in heart failure,6 26 it is plausible that treatment with DITPA reverses abnormalities of calcium handling. The interaction between DITPA treatment and infarction in this study suggests the reversal of a process specific to heart failure.
Thyroid hormone, as long-term treatment for heart failure after myocardial infarction in rats, does not decrease left ventricular end-diastolic pressure at doses that improve left ventricular performance.11 In the present study, DITPA decreased left ventricular end-diastolic pressure while improving ventricular performance. The reasons for the differential effects of the native hormone versus the analogue are not clear. One potential explanation is that thyroxine, at doses necessary for effects on systolic function, produces greater increases in blood volume and venoconstriction than DITPA.27 The combined effect of these changes in the peripheral circulation has been shown to be associated with an increase in left ventricular end-diastolic pressure and volume in rats after myocardial infarction.4 An alternative reason for the decrease in left ventricular filling pressure in infarcted rabbits after DITPA treatment may be related to the marked improvement in ventricular relaxation that was associated with treatment. However, we know of no previous investigations that have examined the independent effects of changes in left ventricular relaxation on ventricular filling pressure, and thus the plausibility of this explanation will need to be established by further studies.
Treatment with DITPA produced no discernible effects on heart weights or on any variable of left ventricular remodeling in either sham-operated or infarcted rabbits. These results are similar to those obtained in our earlier study in rats with large infarctions, where treatment with DITPA and captopril had minimal effects on left ventricular remodeling.15 Earlier studies of infarcted rats treated with captopril have demonstrated attenuation of left ventricular remodeling only in the presence of moderate-sized infarcts.28 It is possible, therefore, that the addition of captopril to DITPA in rabbits with moderate-sized infarctions may have further improved left ventricular performance.
Other potential mechanisms that might explain the action of DITPA were not explored in the present study. Future investigations should include evaluation of the relation between the physiological effects of thyroid hormone analogues such as DITPA and the biochemical and molecular pathways that influence calcium handling in the failing heart.
Although it has been shown in rats after infarction that V/VW is proportional to infarct size,2 we did not obtain sufficient numbers of infarcted rabbits with large (>30%) myocardial infarctions to test the hypothesis that wall thinning in the noninfarcted myocardium is also a function of infarct size in rabbits.
An example of the problems associated with anesthesia can be seen in comparing data in Tables 1⇑ and 2⇑. The resting left ventricular end-diastolic pressure in the conscious infarcted animals is decreased with DITPA treatment (Table 1⇑), whereas with the open chest measurements, there is only a trend for the end-diastolic pressure to be lowered by DITPA (Table 2⇑). In this report, we have stressed the conscious hemodynamics because of the well known limitations of obtaining data during surgery and anesthesia.
Summary and Implications
Rabbits that survive occlusion of the circumflex artery develop myocardial dysfunction and left ventricular remodeling. Along with a previous investigation in which alterations in neurohormones and baroreflex responses were described,20 the present study establishes the rabbit as an alternative model of chronic left ventricular dysfunction. Treatment with the thyroid hormone analogue DITPA produces improvement in ventricular performance and reduces end-diastolic pressure. Whether DITPA represents the prototype of a previously undescribed class of agents29 that will be efficacious for the treatment of heart failure will need to be verified by clinical trials.
This study was supported by grants from the Veterans Administration, the National Institute of Health (Program Project HL-20984 and RO1 HL-48163), Arizona Disease Control Research Commission (82-0697), and the Arizona Affiliate of the American Heart Association. With technical assistance from Alice McArthur.
- Received August 2, 1994.
- Accepted August 31, 1994.
- Copyright © 1995 by American Heart Association
Pfeffer MA, Pfeffer JM, Fishbein M, Fletcher PJ, Spadaro J, Kloner R, Braunwald E. Myocardial infarct size and ventricular function in rats. Circ Res. 1979;44:503-512.
Fletcher PJ, Pfeffer JM, Pfeffer MA, Braunwald E. Left ventricular diastolic pressure-volume relations in rats with healed myocardial infarction: effects on systolic function. Circ Res. 1981;49:618-626.
Raya TE, Gay RG, Lancaster L, Aguirre M, Moffett C, Goldman S. Serial changes in left ventricular relaxation and chamber stiffness after large myocardial infarction in rats. Circulation. 1988;77:1424-1431.
Raya TE, Gay RG, Aguirre M, Goldman S. Importance of venodilatation in prevention of left ventricular dilatation after chronic large myocardial infarction in rats: a comparison of captopril and hydralazine. Circ Res. 1989;64:330-337.
Hodsman GP, Kohzuki M, Howes LG, Sumithran E, Tsunoda K, Johnston CI. Neurohumoral responses to chronic myocardial infarction in rats. Circulation. 1988;78:376-381.
Afzal N, Dhalla N. Differential changes in left and right ventricular SR calcium transport in congestive heart failure. Am J Physiol. 1992;262:H868-H874.
Pfeffer MA, Pfeffer JM, Steinberg C, Finn P. Survival after an experimental myocardial infarction: beneficial effects of long-term therapy with captopril. Circulation. 1985;72:406-412.
Mercadier JJ, Lompre AM, Duc P, Boheler KR, Fraysse JB, Wisnewsky C, Allen PD, Komajada M, Schwartz K. Altered sarcoplasmic reticulum Ca2+-ATPase gene expression in the human ventricle during end-stage heart failure. J Clin Invest. 1990;85:305-309.
Gay RG, Graham S, Aquirre M, Goldman S, Morkin E. Effects of 10-12 days treatment with L-thyroxine in rats with myocardial infarction. Am J Physiol. 1988;255:H801-H806.
Evans R. The steroid and thyroid hormone receptor superfamily. Science (Wash DC). 1988;240:889-895.
Pennock GD, Raya TE, Bahl JJ, Goldman S, Morkin E. Cardiac effects of 3,5-diiodothyropropionic acid, a thyroid analogue with inotropic selectivity. J Pharmacol Exp Ther. 1992;263:163-169.
Pennock GD, Raya TE, Bahl JJ, Goldman S, Morkin E. Combination treatment with captopril and the thyroid hormone analogue 3,5-diiodothyropropionic acid (DITPA): a new approach to improving left ventricular performance in heart failure. Circulation. 1993;88:1289-1298.
Arai M, Otsu K, MacLennan DH, Alpert NR, Periasamy M. Effect of thyroid hormone on the expression of mRNA encoding sarcoplasmic reticulum proteins. Circ Res. 1991;69:266-276.
Shiki K, Hearse DJ. Preconditioning of ischemic myocardium: reperfusion-induced arrhythmias. Am J Physiol. 1987;253: H1470-H1476.
Esser KA, Boluyt MO, White TP. Separation of cardiac myosin heavy chains by gradient SDS-PAGE. Am J Physiol. 1988;255:H659-H663.
Klein MD, Herman MW, Gorlin RG. A hemodynamic study of left ventricular aneurysm. Circulation. 1967;35:614-630.
Pfeffer MA, Braunwald E. Ventricular remodeling after myocardial infarction: experimental observations and clinical implications. Circulation. 1990;81:1161-1172.
Buttrick P, Perla C, Malhotra A, Geenen D, Lahorra M, Scheuer J. Effects of chronic dobutamine on cardiac mechanics and bio-chemistry after myocardial infarction in rats. Am J Physiol. 1991;260:H473-H479.
Nagai R, Zarain-Herzberg A, Brandl CJ, Fujii J, Tada M, MacLennan DH, Alpert NR, Perisamy M. Regulation of myocardial Ca2+-ATPase and phospholamban mRNA expression in response to pressure overload and thyroid hormone. Proc Natl Acad Sci U S A. 1989;86:2966-2970.
Rohrer D, Dillmann W. Thyroid hormone markedly increases the mRNA coding for sarcoplasmic reticulum Ca2+-ATPase in the rat heart. J Biol Chem. 1988;263:6941-6944.
Takahashi T, Allen PD, Marks AR, Denniss AR, Schoen FJ, Grossman W, Marsh JD, Izumo S. Altered expression of genes encoding the Ca2+ regulatory proteins in the myocardium of patients with end-stage heart failure: correlation with expression of the Ca2+-ATPase gene. Circ Res. 1992;71:1357-1365.
Gay RG, Raya TE, Lancaster LD, Lee RW, Morkin E, Goldman S. Effects of thyroid state on venous compliance and left ventricular performance in rats. Am J Physiol. 1988;254:H81-H89.
Pfeffer JM, Pfeffer MA, Braunwald E. Influence of chronic captopril therapy on the infarcted left ventricle of the rat. Circ Res. 1985;57:84-95.