Journal Logo


Long-Term Effects of Carvedilol on Left Ventricular Function, Remodeling, and Expression of Cardiac Cytokines After Large Myocardial Infarction in the Rat

Sia, Ying Tung*; Parker, Thomas G.; Tsoporis, James N.; Liu, Peter; Adam, Albert; Rouleau, Jean L.

Author Information
Journal of Cardiovascular Pharmacology: January 2002 - Volume 39 - Issue 1 - p 73-87
  • Free


Morbidity and mortality after large myocardial infarction (MI) remain elevated despite significant advances in therapy over the past decade (1). The mechanisms of death in this setting include recurrent ischemia (2), progressive heart failure (3), and ventricular arrhythmias (1,4). These three are interrelated and are associated with progressive post-MI ventricular dilatation (5). Angiotensin-converting enzyme inhibitors have been shown to reduce morbidity and mortality after large MI in both animals (6) and humans (7). One of the major mechanisms by which they do this is by reversing adverse post-MI ventricular remodeling and by attenuating neurohormonal activation. β-Blockers have also been shown to reduce post–large MI morbidity and mortality (8,9). The mechanisms by which β-blockers exert their beneficial effects appear to be complex and involve improved ventricular remodeling, preserved cardiac contractility and adrenergic responsiveness, reduced oxidative stress, and reduced cardiac expression of proinflammatory cytokines (10,11). However, controversy as to the effects of β-blockers on cardiac remodeling remains, as we demonstrated no benefit of propranolol on post-MI ventricular dilatation (12) and Hu et al. (13) found no benefit of bisoprolol on post-MI remodeling in the rat post-MI model. Few clinical data exist on the long-term post-MI ventricular remodeling effects of β-blockers.

Carvedilol is a β 12 -blocker that has α 1 -blocker and anti-oxidant properties (14). It has been shown to improve ventricular function (15), reduce left ventricular dilatation (15), and improve survival in patients with post-MI left ventricular dysfunction (9) or chronic heart failure caused by ischemia (15,16). Because of its α 1 -blocker vasodilatory properties, it is conceivable that carvedilol could have beneficial effects on ventricular remodeling that other β-blockers do not. Indeed, studies by Khaper and Singal (17) using the α 1 -blocker prazosin in this post-MI model indicate that the long-term use of prazosin results in improved left ventricular hemodynamics. Also, carvedilol may have beneficial effects related to its anti-oxidant properties. Oxidative stress could contribute to poor post-MI prognosis (18) via numerous mechanisms of which direct cytotoxic (19,20), negative inotropic (19–21), cytokine-stimulating (22), adrenergic system–uncoupling (23), adverse ventricular remodeling (24), and apoptotic effects (25,26) are but a few. In favor of potential beneficial effects of carvedilol over other β-blockers is a report by Wei et al. (27) demonstrating that carvedilol has beneficial effects on cardiac fibrosis in the post-MI setting that metoprolol does not. However, at this time, few data exist regarding the effects of carvedilol after MI.

In this study we tested the hypothesis that in large MI (> 45% of the circumference of the left ventricle) carvedilol leads to beneficial post-MI left ventricular remodeling and that this is associated with improved survival in the rat post-MI model. In addition, we evaluated the effects of carvedilol on cardiac hemodynamics, on intrinsic myocardial contractility and β-adrenergic responsiveness, and on cardiac expression of inflammatory cytokines and oxidative stress.


Animals' care

Male Wistar rats weighing 200–250 g were obtained from Charles River Breeding Laboratories (Saint-Constant, Quebec, Canada). All care was in accordance with the Canadian Council for Animal Care and the Animal Care Committee guidelines of the Montreal Heart Institute.

Myocardial infarction operative procedure

MI was induced in rats by ligating the left anterior coronary artery as described by Pfeffer et al. (6) There was a high early mortality rate, with 51% of deaths occurring within 24 h of infarction. The survivors were randomly allocated to the two treatment groups. They were not classified according to MI size until the end of the study, at which time they were separated into sham, small to medium, or large MI according to criteria described by Nguyen et al. (28). Rats with a large MI were defined as those having a left ventircular scar of > 45% of the left ventricular circumference or a scar-to-body-weight ratio > 0.2 g/kg. Those classified as having small to medium MI (left ventricular scar < 45% of the circumference or scar-to-body-weight ratio < 0.2 g/kg) were not further considered. This classification is based on a relationship between MI circumference, scar-to-body-weight-ratio, and left ventricular end-diastolic pressure (LVEDP) (28). Rats that died > 72 h after coronary ligation but prior to hemodynamic monitoring also had morphologic assessment for classification of MI size. Those dying between 24–72 h after infarction were assumed to have had a large MI. Sham-operated animals had the same operative procedure except that the suture was not tied and thus the coronary artery was not ligated.

Pharmacologic interventions

Rats surviving 24 h after MI were randomized into three groups as follow: vehicle (soya bean oil); carvedilol (20 mg/kg b.i.d.) (29) given continuously until the time of hemodynamic monitoring (“carvedilol cont'd med” group); or carvedilol (20 mg/kg b.i.d.) with treatment interrupted 24–48 h prior to the time of hemodynamic monitoring (carvedilol washout group). Rats received their drugs by gavage. Carvedilol was homogenized in soya bean oil. All rats were treated for 28–30 days (Fig. 1).

FIG. 1.
FIG. 1.:
Various groups of rats according to myocardial infarction (MI) size and treatment group. All rats were randomized 20–24 h after coronary artery ligation. MI size was determined at the end of the study, and thereafter the rats were classified according to their MI size.

Long-term cardiac hemodynamic studies

After 4 weeks of treatment, the rats were initially anesthetized with 3% halothane mixed with 100% O 2 . The halothane percentage was reduced to 1% 3 min prior to hemodynamic recordings. For left ventricular pressure measurements, the right carotid artery was isolated and a 2F microtip pressure transducer catheter (model SPR-407, Millar Instrument Inc., Houston, TX, U.S.A.) was inserted and then advanced into the left ventricle to measure left ventricular systolic pressure (LVSP), LVEDP, and heart rate. Left ventricular +dP/dt and −dP/dt (maximum rate of pressure rise [+] or decline [−]) were derived by active analogue differentiation of the pressure signal. Similarly, for right ventricular pressure measurements, a 2F Millar catheter with the pressure sensor at the tip deviated from the shaft by an angle of about 130° was inserted into the right jugular vein. It was then advanced into the right ventricle, and right ventricular systolic pressure (RVSP), right ventricular end-diastolic pressure (RVEDP), and right ventricular +dP/dt and −dP/dt were obtained. All tracings were recorded on a Gould 2600-S recorder (Gould, Cleveland, OH, U.S.A.).

Once hemodynamic monitoring was done, papillary muscle contractile characteristics, cardiac morphology, and left ventricular remodeling were assessed (Fig. 1).

Isolated papillary muscle studies

Once cardiac hemodynamics were completed, 14 rats were killed by cardiopulmonary excision under heavy anesthesia with halothane. The heart was rapidly separated from the lung and washed in Krebs-Henseleit solution. One or two left ventricular papillary muscles were excised immediately from each heart. Papillary muscles were then mounted in a bath with Krebs-Henseleit solution containing NaCl 118 m M, KCl 3.5 m M, MgSO 4 2.43 m M, CaCl 2 0.7 m M, KH 2 PO 4 1.2 m M, NaHCO 3 24.9 m M, and dextrose 5.0 m M and bubbled with a 95% of O 2 and 5% of CO 2 mixture at a pH of 7.4. The base of the muscle was held by a stainless steel clamp and the other end was tied to a lever with an electromagnetic feedback system identical to that described by Li et al. (30). The muscles were stimulated at 10% above threshold at six stimuli per minute with a model S-88 stimulator (Grass Instrument Co., Quincy, MA, U.S.A.) through platinum field electrodes. The preload was adjusted so that the muscle length was at Lmax (the length at which maximum developed tension occurred). The muscles were then allowed to stabilize for 2 h prior to contractile measurements. Contractions were recorded at 100 mm/s on a recorder (Brush 200, Gould). The output of the Gould recorder was connected to an electronic differentiator and both of these were fed into an A/D converter (DATA/translation, DT 2821-F801; Marlborough, MA, U.S.A.), which in turn was connected to a COMPAQ 486 (Houston, TX, U.S.A.) microcomputer. Analysis of force characteristics was performed by custom software running under MS-DOS. Muscle cross-section was assessed by dividing muscle weight by muscle length, assuming a cylindrical shape and a specific gravity of 1. For carvedilol-treated rats, only papillary muscles from continuously treated rats were used for these studies.

After a 2-h stabilization period at Lmax, baseline isometric and unloaded (zero load for Vmax) contractions were recorded. Cumulative concentration-response curves for isoproterenol from 10 −8M–3 × 10 −5M were performed by repeating the same measurements after a 10-min stabilization period at each concentration. Once the dose-response curves for isoproterenol were done, two concentrations of calcium (2.25 m M and 10 m M) were given to assess maximal contractile capacity. Again, isometric and unload contractions were recorded at each concentration.

Passive pressure–volume relationship

After completing the cardiac hemodynamic measurements, 58 rats had their hearts stopped in diastole by an i.v. injection of a saturated potassium chloride solution via the right jugular vein. The heart was then removed and rinsed with a saline solution. Immediately after the rinse, the right ventricle was opened to eliminate its influence on left ventricular expansion. Then a double-lumen catheter (Becton Dickinson; Parsippany, NJ, U.S.A.) (PE-50 inside PE-200) was inserted 6 mm into the left ventricle via the aorta, the atrioventricular groove was ligated, and the left ventricle was emptied of all remaining solution. A negative pressure (−5 mm Hg) was achieved in the left ventricle using a syringe connected to a three-way stopcock positioned between the left ventricle and the pressure transducer. Physiological saline was infused at 0.68 ml/min via the PE-200. This was done for pressures between −5–35 mm Hg. Three pressure–volume curves were obtained within 10 min and an average of these three curves was used as the final value.

Cardiac remodeling

Once the pressure–volume curve was completed, the left ventricle was filled with saline solution to a pressure of 15 mm Hg, sealed, and fixed in its distended form in 10% formalin phosphate buffer for 24 h. After fixation, the atria and great vessels were dissected, the right ventricle was cut off the heart along its septal insertion, and the right and left atria and right and left ventricular were weighed separately. Two cross-sections were obtained at 1-mm intervals midway between the base and the apex of the left ventricle.

Morphometry of the left ventricle was performed on tissues obtained from the two cross-sections. These tissues were dehydrated at room temperature through graded ethanol series and embedded in parafin. Transverse sections of 4-μm thick were cut and two cross-sections of each left ventricle were stained with hematoxylin-phloxin-safran and mounted on a slide for projection to a magnification × 10 under a Zeiss-Jena magnifier (Imaging Products International, Inc., Chantilly, VA, U.S.A.). Sections were traced on a calibrated digitizing tablet and morphologic variables calculated directly by computerized planimetry with Sigma Scan software (Labtronics Inc., Guelph, Ontario, Canada). The length of the scar and that of noninfarcted muscle for epicardium and endocardium were numerically summed separately. The epicardial and endocardial circumferences were obtained directly by planimetry. The ratio of the sum of the length of the scar and the circumference defined the MI size of each of the myocardial surfaces. Final infarct size was expressed in percent, as the average of the infarct size of the endocardial and epicardial surfaces × 100. The average scar thickness was obtained directly by planimetry as well. The left ventricular surface area at the endocardium (cavity) and epicardium (cavity and ventricular wall = ventricular area) were numerically summed separately. For both cross-sectional levels, dilatation index was expressed as the ratio of left ventricular cavity area/left ventricular area. Data from the “carvedilol cont'd med” group and carvedilol washout group were similar for these measurements and were thus considered together.

Cardiac fibrosis assessment

This procedure consisted of using samples from both cross-sections (12). These were cut into 8-μm-thick slices and stained with Sirius red F3BA (BOH Laboratory Supply; Poole, England) as a 0.1% solution in saturated aqueous picric acid. The Sirius red staining technique has a good affinity for collagen fibers and gives pictures of high contrast using polarized light, as described previously by Nguyen et al. (31) Each slide was examined under a Zeiss-Axioskop microscope (Imaging Products International) fitted with cross-polarizing filters at a final magnification × 200, giving a final resolution of 0.72 μm 2 per pixel. The collagen network was quantified by computer-assisted image analysis. The automated system included a digital image analysis device Samba 4000 (Imaging Products International) based on morphologic mathematic principles connected to an IBM-compatible personal computer running customized algorithms written in C language. Each microscopic field of observation sent to the image analyzer was transmitted by a Sony 3CCD video camera (Scott Scientific; Montreal, Quebec, Canada) and transformed into a binary image. Then, a sequence of mathematical and morphologic operations allowed us to identify and quantify the collagen network. Collagen volume density fraction was then determined by measuring the area of stained tissue within a given field and expressing that area as a proportion of the total area under observation. The areas studied were far from the scar, and the collagen-rich border zone of vessels was not included in the calculations. Ten fields were analyzed in the subendocardial layer and 10 fields in the subepicardial layer in each left ventricle. An approximation of total collagen volume was obtained by multiplying left ventricular weight to body weight ratio by collagen volume density percent.

Cardiac and lung weights

A total of 229 rats were used to assess cardiac hypertrophy. Once the hemodynamic protocol was completed, rats were killed by cardiopulmonary excision. The heart was rapidly separated from the lung, washed in Krebs-Henseleit solution, and dissected into right and left atria, right ventricle, left ventricle (including septa), and scar. The scarred area was pinned on a paper and its surface was determined by planimetry. Each tissue was then weighed individually, frozen in liquid nitrogen, and stored at −80°C. The lungs were also weighed and frozen. Data from the “carvedilol cont'd med” group and the carvedilol washout group were similar and were thus considered together.

Plasma norepinephrine measurements

Blood was obtained just before sacrifice, and from it plasma norepinephrine was measured by methods described previously (12).

Quantification of fetal gene expression in left ventricular myocardium

Myocardial fetal gene expression, as assessed by measuring mRNA for atrial natriuretic factor, β myosin heavy chain (βMHC), α myosin heavy chain (αMHC), and skeleton actin (skACT), was evaluated in myocardium from the left ventricle of six hearts from each group. The methods used have previously been described (32). Briefly, RNA was isolated from tissues by a one-step acid guanidinium phenol method. RNase protection assays to determine steady-state levels of skACT mRNA, atrial natriuretic factor mRNA, βMHC/αMHC mRNAs, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNAs were performed by modifying conditions previously described. Anti-sense riboprobes for the rat skACT (a gift from Dr. L.K. Karns, Laboratory of Molecular Neuro-Oncology, University of Virginia Health Sciences, Charlottesville, VA, U.S.A.), mouse/rat atrial natriuretic factor (33), mouse/rat βMHC/αMHC (34) (a gift from Dr. C.S. Long, Division of Cardiology and Research, Veterans Affairs Medical Center, San Francisco, CA, U.S.A.), and GAPDH (Ambion, Austin, TX, U.S.A.) were labeled with a [ 32 ]UTP (uridine triphosphatase) (800 Ci/mmol, Amersham, Oakville, Ontario) by in vitro transcription with T7 (skACT, βMHC/αMHC) or SP6 (GAPDH, S100β, atrial natriuretic factor) RNA polymerase of appropriate RNA synthesis vectors, respectively. The probes (2 ng, 10 6 disintegration per min) were added to 15 mg of total RNA from rat tissues, or rat tRNA, in 20 ml of 300 m M Na acetate, pH 6.4, containing 100 m M Na citrate, 80 m M formamide, and 1 m M ethylenediaminetetra-acetic acide. The hybridization reaction was incubated for 18 h at 45°C and the annealed products were digested with RNase A (1 unit/ml) and RNase T1 (100 units/ml) at 37°C for 30 min. RNase-resistant hybrids were recovered using a commercial kit (Ambion) and analyzed on 8 M urea-6% polyacrylamide sequencing gels and visualized by autoradiography.

Ribonuclease protection assay for cardiac inflammatory cytokine panel

To determine the gene expression of cytokines from the affected myocardium, ribonuclease protection assay (RPA) was performed to quantitate the mRNA levels of tumor necrosis factor-α and interleukin-1β, -5, and -6. The cDNA probes specific for the rat target cytokines were prepared from a commercially available kit based on known coding sequences (RiboQuant, PharMingen; San Diego, CA, U.S.A.), according to manufacturer's instructions. Specifically, the cDNA templates of the target cytokines were amplified using polymerase chain reaction and labeled with α 33 P-UTP (Dupont, New England Nuclear; Boston, MA, U.S.A.). Commercially available cDNA probes for housekeeping genes L32 and GAPDH were also amplified and labeled to act as standardization controls.

The total RNA from the myocardial samples was extracted and purified according to techniques previously published from our laboratory (35), taking particular care to ensure Rnase-free reaction conditions. RPA was performed with the co-addition of 10 μg of total RNA, the labeled probes (at least 2 × 10 4 cpm) and yeast tRNA to act as background control into prepared hybridization buffer, incubated for 16 h at 53°C. RNase digestion was then performed with RNase A and RNase T1 at 30°C for 45 min. The reaction was terminated with proteinase K, followed by phenol-chloroform extraction and ethanol precipitation. The protected fragments were then resolved on 5% polyacrylamide gel electrophoresis together with standard size markers. After gel drying and autoradiography, the probe activities were documented by phosphoimager (Biorad; Mississawga, Ontario, Canada) and the results digitally quantitated and normalized to GAPDH for each sample at each time point observed with National Institutes of Health Image software.

Cardiac oxidative stress by gas chromatography: mass spectrometry

Cardiac oxidative stress was assessed by measuring the aldehyde compounds and their byproducts categorized into four groups: saturated aldehydes; aldehydes with double bonds; hydroxyl-aldehydes with double bonds; and malondialdehydes. Samples were analyzed by a modification of the previously described method from Luo et al. (36). Briefly, heart tissue (100 mg of tissue) was homogenized in 1 ml of deionized filtered water containing Butylated hydroxytoluene and deferoxamine mesylate at a final concentration of 20 μM. A fixed amount of benzaldehyde-ring-D 5 (100 pmol) was added to heart homogenates as internal standard. A 0.05-M solution (200 μl) of pentafluorobenzyl hydroxylamine hydrochloride (PFBHA HCl) was added to each sample to form aldehyde-PFBHA derivative. The mixture was vortex mixed and incubated for 30 min at room temperature. After incubation, samples were deproteinized with the addition of 1 ml of methanol. Aldehyde-PFBHA derivatives were extracted with 1 ml of hexane. The hexane phase was collected and the extraction step was repeated. The combined hexane layers were dried over 0.5 g of anhydrous sodium sulfate. The hexane solvent was evaporated under a stream of nitrogen gas and 50 μl of N, O-bis (trimethylsilyl) trifluoroacetamide (BSTFA) +1% trimethylchlorosilane (TMCS) was added and incubated at 60°C for 15 min to form TMS ether derivatives of the hydroxyl functional groups. Finally, 50 μl of hexane was added to each sample and analyzed by gas chromatography mass spectrometry.

The PFB-oxime-TMS derivatives were analyzed by capillary column gas-chromatography-negative-ion chemical ionization mass spectrometry (GC-NICIMS) with methane as reagent gas. GC-NICIMS was performed using the bench-top Agilent Technologies 5973 MSD GC/MS system (Agilent Technologies Inc.; Palo Alto, CA, U.S.A.). The aldehyde separation was conducted on a 30 m, 0.25 mm internal diameter, 0.25-μm-film-thickness Hewlett Packer-5ms column (Agilent Technologies Inc.). The GC conditions were as follows: the carrier gas was helium at a flow rate of 1.2 ml/min; samples were injected by pulsed splitless injection; the injection port temperature was 250°C, and the transfer line was kept at 280°C. A specific ion for each derivative was selected for selected ion monitoring.

Statistical analyses

All values are expressed as mean ± SEM. Results were analyzed by using a two-tailed Student's t test for unpaired data and by analysis of variance for multiple comparisons, followed by a two-sided Dunnett test when appropriate. Statistical significance was assumed at p < 0.05. Kaplan-Meier survival curves over the follow-up period were constructed and analyzed by the generalized Savage (Mantel-Cox) test. Gels were analyzed by densitometry. Results are presented as mean (arbitrary units) ± SEM. Mean values were compared with the left ventricular sham MI group by analysis of variance, followed by Student-Newman-Keuls test, with significance of p < 0.05 respective to the left ventricular sham MI group.



A total of 843 rats were used in the study. Of these, 434 died within 24 h after coronary ligation and thus were excluded from the study. The remaining 409 rats were treated randomly with soya bean oil (n = 201) or carvedilol (n = 208). Regardless of treatment group, the survival rate of sham rats was excellent over the 4 weeks of follow-up.

The early (< 72 h) post-MI mortality of carvedilol-treated rats with a large MI was higher than that of the soya bean oil group (47 deaths; 27.6%) versus 26 deaths (16.9%), respectively (p = 0.001). When the survival analysis was performed after 72 h after MI, the survival rates for both soya bean oil and carvedilol groups were similar (60.9% versus 53.7%). The overall survival rate (28 days) of rats with a large MI was 50.6% for soya bean oil and 38.8% for carvedilol (soya bean oil versus carvedilol, p = 0.013).

Hemodynamic measurements

In the sham rats, carvedilol treatment (“cont'd med”) had no effect on any of the hemodynamic parameters measured, except for heart rate, which it decreased significantly as compared with soya bean oil (vehicle) (Table 1). Discontinuing carvedilol for 24–48 h before performing the hemodynamic measurements resulted in a normalization of heart rate and an increase in LVSP.

Hemodynamic monitoring 4 weeks after myocardial infarction (MI)

Vehicle rats with a large MI had a decrease in heart rate, LVSP and left ventricular +dP/dt and −dP/dt as compared with their sham counterparts. This was accompanied by an increase in LVEDP, RVSP, RVEDP, and right ventricular +dP/dt. As compared with vehicle rats with a large MI, “carvedilol cont'd” treatment reduced heart rate but resulted in no other difference in hemodynamic measurements. After carvedilol was interrupted for 24–48 h (washout), heart rate returned to control levels and there were significant increases in LVSP, LVEDP, and left ventricular +dP/dt as compared with both the vehicle and “carvedilol cont'd” large MI groups. Their right ventricular +dP/dt was significantly increased as compared with the carvedilol (“cont'd med”) group.

Isolated papillary muscle study

At baseline, papillary muscle contractile characteristics from sham rats were similar, regardless of treatment group (Table 2). Papillary muscles from the vehicle large MI group had significant decreases in total tension, maximum rate of tension rise and decline (+dP/dt and −dP/dt), and Vmax as compared with the vehicle sham group. As compared with the vehicle large MI group, the carvedilol (“cont'd”) large MI group had an increase in total tension, +dP/dt and Vmax, suggesting preserved myocardial contractility.

Isolated papillary muscle studies 4 weeks after myocardial infarction (MI)

The effects of a dose of 3 × 10 −5 of isoproterenol were similar in all of the sham groups, indicating at least partial carvedilol washout in the isolated muscle bath. In sham-operated rats, as has been well described before (37), tension increased only slightly with isoproterenol due to the near saturating intracellular calcium concentrations under basal conditions, but +dP/dt increased markedly. In all large MI groups, isoproterenol resulted in a paradoxical decrease in total tension, this decrease (Δ%) tending to be less in the “carvedilol cont'd” group. The increases in Vmax, +dP/dt, and −dP/dt with isoproterenol in the vehicle large MI group were also less than in their sham counterparts. Again “carvedilol cont'd” tended to normalize the response to isoproterenol, suggesting that they had a protective effect on β-adrenergic signal transduction. The addition of high-dose extracellular calcium had similar effects in all three (vehicle and two carvedilol groups) large MI groups.

Morphologic assessment

There was no difference in morphologic characteristics between the two sham groups (Tables 3 and 4, Fig. 2).

Morphology parameters of heart 4 weeks after myocardial infarction (MI)
Morphologic characteristics of mid left ventricular cross-sections 4 weeks after myocardial infarction (MI)
FIG. 2.
FIG. 2.:
Representative left ventricular midventricular cross-sections of a normal rat (A) and rats with large MI treated with vehicle (B) or carvedilol (C).

The two groups with a large MI had a significant decrease in body weight. As expected, scar weight and scar weight/body weight ratios were similar in both large MI groups. Atria weight/body weight ratio increased similarly in both large MI groups, but differences were found in the left ventricular weight/body weight ratio. The left ventricular weight/body weight ratio was greater in the carvedilol large MI group as compared with the vehicle large MI group.

In rats with a large MI, there was an increase in left ventricular endocardial and epicardial circumferences, in scar surface, cardiac fibrosis, and dilatation index as compared with the sham groups (Fig. 2, Table 4). Carvedilol-treated rats with a large MI had an increase in scar thickness but a decrease in cardiac fibrosis as compared with vehicle-treated rats with a large MI. Left ventricular epicardial circumference was similar in the two large MI groups. However, consistent with their increase in left ventricular weight, carvedilol-treated rats with a large MI also had a decrease in dilatation index, which is a measure of the index of left ventricular hypertrophy to dilatation, as compared with vehicle rats with a large MI.

Passive pressure–volume relationship

In sham rats, passive pressure–volume relationships (Fig. 3) were similar regardless of treatment group. As expected, a large MI caused a rightward shift of this relationship as compared with sham groups. Treatment with carvedilol did not modify this rightward shift.

FIG. 3.
FIG. 3.:
Passive left ventricular pressure–volume relations of rats with sham to small myocardial infarction (MI) and large MI according to treatment group.

Plasma norepinephrine 4 weeks after myocardial infarction

In the sham groups, plasma norepinephrine levels were similar, 505 ± 62 pg/ml and 671 ± 162 pg/ml for vehicle and carvedilol, respectively. In the large MI vehicle group, there was a significant increase in norepinephrine levels (1,183 ± 211 pg/ml, p < 0.01). “Carvedilol cont'd” treatment had no effect on this increase (1,036 ± 121 pg/ml).

Fetal gene expression 4 weeks after myocardial infarction in the left ventricle

Under sham conditions, expression of skACT, atrial natriuretic factor, and βMHC were barely detectable in left ventricular tissue, whereas αMHC showed marked expression (Table 5, Fig. 4). In the vehicle post-MI group, a five- to six-fold induction of skACT, atrial natriuretic factor, and βMHC was observed, whereas αMHC mRNA levels were reduced to about 50% of those under sham conditions. Treatment with carvedilol (“cont'd med”) had no effect on gene expression under sham conditions or in large MI. GAPDH steady-state mRNA levels served as a control for the quality and loading of the mRNA.

Fetal gene expression 4 weeks after myocardial infarction (MI) in the left ventricle
FIG. 4.
FIG. 4.:
Representative Rnase protection assay showing steady-state levels of atrial natriuretic factor (ANF), skeleton actin (skACT), α and β myosin heavy chain (MHC), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNAs in rat adult brain, fetal left ventricular (LV), or left ventricular free wall under sham to small myocardial infarction (MI) conditions, and after MI (left anterior descending artery) following treatment with either vehicle (soya bean oil) or carvedilol (Carv). Protected fragments specific for rat ANF, skACT, αMHC/βMHC, and GAPDH mRNAs are indicated. Specificity of each band is indicated by the absence of signal with tissue RNA.

Cardiac inflammatory cytokine gene expression

There were no differences in the myocardial expression of cytokines in the two sham groups (Fig. 5). The vehicle MI group had an increase in only interleukin-1β cardiac expression as compared with the vehicle sham group. The carvedilol MI group had a decrease in only cardiac interleukin-1β expression as compared with the vehicle MI group.

FIG. 5.
FIG. 5.:
Myocardial expression of inflammatory cytokines of rats with sham to small and large myocardial infarction (MI) according to treatment group: (A) tumor necrosis factor-α (TNF-α), (B) interleukin-β (IL-β), (C) IL-5, and (D) IL-6.

Cardiac oxidative stress

There were no differences in the myocardial oxidative stress markers between the two sham groups (Fig. 6). The vehicle large MI group had an increase in only myocardial malondialdehydes as compared with its control counterparts. Carvedilol treatment of rats with large MI had no effect on this increase compared with the vehicle large MI group.

FIG. 6.
FIG. 6.:
Myocardial oxidative stress of rats with sham to small myocardial infarction (MI) and large MI according to treatment group. A: Saturated aldehydes. B: Aldehydes with double bonds. C: Hydroxyl-aldehydes with double bonds. D: Malondialdehydes.


This study indicates that over a relatively short period (28 days) following a large MI, carvedilol results in preservation of myocardial contractility and β-adrenergic signal transduction. It also has beneficial effects on ventricular remodeling and reduces the cardiac expression of the inflammatory cytokine interleukin-1β but not cardiac oxidative stress. Despite these beneficial effects, carvedilol reduced survival early (4 weeks) after MI, perhaps due to the use of excessive doses very early after MI. Further studies using progressively increasing doses of carvedilol after MI and with a longer follow-up are warranted.

In the rat model of large MI, carvedilol treatment resulted in preservation of left ventricular systolic function. Rats with a large MI in which carvedilol was discontinued prior to sacrifice had evidence of preserved left ventricular systolic function. Even rats in which carvedilol was continued until they were killed had cardiac function similar to that of untreated MI controls despite significant β-blockade. Isolated papillary muscle studies clearly indicate that part of this beneficial effect on ventricular function was the result of preserved intrinsic myocardial contractility, which may have been partially explained by reduced cardiac expression of the negative inotropic cytokine interleukin-1β but not by inhibition of cardiac oxidative stress. Other contributing factors include β-adrenergic hyperresponsiveness associated with β-blocker withdrawal and normalization of wall stress due to increased wall thickness. Although increased left ventricular wall thickness may have partially explained the increase in LVEDP found with carvedilol withdrawal, because the passive pressure–volume relationship with carvedilol was not altered, it is more likely the result of increased afterload and heart rate associated with the β-adrenergic withdrawal state. The preservation in adrenergic signal transduction with carvedilol is compatible with the findings of Prabhu et al. (10) with metoprolol.

Although carvedilol did not modify the pressure–volume relationship, it did result in significant morphologic changes such as a thicker scar, decreased cardiac fibrosis, an increase in left ventricular/body weight ratio, and a decrease in cardiac dilatation index. Although our results with cardiac fibrosis are compatible with those of Wei et al. (27), they contrast with those we reported with the nonselective β-blocker propranolol, for which we found excessive cardiac fibrosis (12). These differences may be the result of a decrease in cardiac cytokine expression that would be expected to reduce cardiac fibrosis (38). Apparently, the reported anti-oxidant effects of carvedilol (29) did not play a role, perhaps because the dose of carvedilol used was not sufficient to provide a significant anti-oxidant property. Considerable variability exists as to the effects of β-blockers on post-MI ventricular remodeling. Prabhu et al. (10) found that metoprolol significantly reduced post-MI ventricular dilatation, but in a previous study with propanolol we found excessive ventricular dilatation and no change in left ventricular/body weight ratio in hearts with a similar size of MI (12), and results by Hu et al. (13), if anything, also showed excessive ventricular dilatation with bisoprolol. Why our present and previous results, and the results of Hu et al., contrast with those of Prabhu et al. (10) remains speculative, but the most likely explanation are differences in the length of follow-up. With a longer follow-up it may be that the multiple beneficial effects of β-blockers described in our study would result in improved ventricular remodeling and function. Compatible with this are the multiple clinical studies showing β-blockers to reduce ventricular dilatation and improve ventricular function in chronic heart failure (15,16).

One surprising result with carvedilol is the increase in left ventricular weight and in left ventricular wall thickness found in hearts with a large MI. Indeed, both β- andbgr;- and α-blockade would be expected to reduce cardiac hypertrophy (32,39) and carvedilol's vasodilator properties would be expected to reduce reactive hypertrophy (40,41), yet the contrary occurred. Interestingly, Hu et al. (13) found a similar tendency in post-MI rats with moderate and large MI treated with bisoprolol. We are at a loss to explain this finding although it could result from reduced apoptosis. Patients who have had a large MI have been found to have apoptosis long after their MI, a development that could contribute to inadequate reactive hypertrophy/ventricular remodeling and eventually to cardiac decompensation (42). The β-blocker metoprolol has been shown to reduce apoptosis in a dog model of heart failure (43). Presumably the mechanisms involved would include a reduction in adrenergic tone, wall stress, and cardiac cytokine expression (44), thus increasing in left ventricular weight despite a reduction in stimuli to hypertrophy.

Moreover, in response to MI, surviving terminally differentiated cardiac myocytes undergo a phenotypic transition after MI, involving reexpression of genes normally restricted to the fetus, including the isoform switch from adult α- toagr;- to embryonic βMHC, the skeletal isoform of α-actin, and atrial natriuretic factor (45). The signaling pathways activated by an MI are complex and the relative contributions of hypoxia, ischemia, and activation of local and systemic trophic factors are still unknown. The results of the current study suggest that modulation of pathways involving β-adrenergic signaling and cytokine expression are not involved in the induction of the fetal genetic phenotype following an MI, although the relatively small number of samples studied (n = 6) renders our results nondefinitive. Myocardial contractile characteristics of papillary muscles from carvedilol-treated hearts with large MI were much less depressed than those of the other two groups, despite similar expression of fetal genes in all three large-MI groups. This suggests that the relationship between fetal gene expression and myocardial contractility is more complex than previously thought (46). Because most of the difference between carvedilol and the other groups resided in tension rather than speed of contraction, i.e., Vmax, it would appear that fetal gene expression modifies speed of contraction more than it does tension generating capacity.

In this study treatment with carvedilol did not result in an improvement in early (4 week) post-MI survival despite causing a number of beneficial changes both in vivo and in vitro. This contrasts with the results of the Carvedilol Post-Infarction Survival Control in Left Ventricular Dysfunction (CAPRICORN) study, which found carvedilol to reduce mortality when started early after MI in patients with left ventricular dysfunction (9). The most likely explanation is that we used an excessively large dose of carvedilol early after MI and that this resulted in excessive hypotension and death. In clinical studies of patients with left ventricular dysfunction and in CAPRICORN, β-blockers are given in increasing doses to prevent excessive hypotension and cardiac depression. This would appear to be particularly important after MI, when excessive hypotension can mask the beneficial effects of angiotensin-converting enzyme inhibitors (the Cooperative New Scandinavian Enala Study or CONSENSUS study) (47). In clinical studies, β-blockers have consistently been shown to improve both early and late survival after MI (8,9). In this experimental rat model, Hu et al. found that bisoprolol improved survival when started early but not later after MI (13). The current study should thus be repeated starting with lower doses of carvedilol to avoid excessive hypotension in the early post-MI period, later increasing to full β-blockade, much as is done in clinical practice (15,16). As suggested from clinical trials and the study by Prabhu et al. (9,10), follow-up should also be more prolonged than in this study to permit all of the beneficial effects of carvedilol to exert their full effects. These beneficial effects of carvedilol include preserved myocardial contractility, reduced wall stress, decreased cardiac fibrosis, and decreased cardiac expression of interleukin-1β.

Our lack of improved survival with carvedilol may also be partially explained by the reduced response of rat myocardium to β-adrenergic stimulation as compared with larger mammals. Rat myocardium works at a near saturating intracellular calcium concentrations such that β-blockade would be expected to have less of an effect (48). Finally, the MI is completed more rapidly in this model than it does in humans, where it can be stuttering, thus reducing the potential beneficial anti-ischemic/ventricular fibrillation effects of β-blockade. Had we not selected only rats with a large MI for analyses in this study, there could have been a selection bias in favor of survivors with carvedilol, and comparisons between groups would have been difficult to make. However, according to protocol, the rats considered in our analyses had similar MI sizes and the effects found with carvedilol were considerable and unlikely related to selection bias. Nevertheless selection bias cannot be entirely ruled out.


Supported by the Medical Research Council of Canada. SmithKline Beecham Pharmaceuticals supplied the carvedilol. The authors thank Quang Nyugen, Hugues Gosselin, Luc Moquin, Tan Nyugen, Elias El Salibi and Chantal St-Cyr for their technical support. Authors are grateful to Dr. Jean-Gilles Latour (University of Montreal) for collagen quantification. We also thank Carolyn Gillis and Diane Levreault for their excellent work in manuscript preparation


1. Rouleau JL, Talajic M, Sussex B, et al. Myocardial infarction patients in the 1990s: their risks factors, stratification and survival in Canada: the Canadian Assessment of Myocardial Infarction (CAMI) Study. J Am Coll Cardiol 1996; 27: 1119–27.
2. Barbagelata A, Granger CB, Topol EJ, et al. Frequency, significance and cost of recurrent ischemia after thrombolytic therapy for acute myocardial infarction. TAMI Study Group. Am J Cardiol 1995; 76: 1007–13.
3. Vaughan DE, Pfeffer MA. Post-myocardial infarction ventricular remodeling animal and human studies. Cardiovasc Drugs Ther 1994; 8: 453–60.
4. Opitz CF, Mitchell GF, Pfeffer MA, et al. Arrhythmias and death after coronary artery occlusion in rat: continuous telemetric ECG monitoring in conscious untreated rats. Circulation 1995; 92: 253–61.
5. Pfeffer MA, Braunwald E. Ventricular remodeling after myocardial infarction: experimental observations and clinical implications. Circulation 1990; 81: 1161–72.
6. Pfeffer JM, Fischer TA, Pfeffer MA. Angiotensin-converting-enzyme inhibition and ventricular remodeling after myocardial infarction. Annu Rev Physiol 1995; 57: 805–26.
7. Pfeffer MA, Braunwald E, Moyé LA, et al. on behalf of the SAVE Investigators. Effect of captopril on mortality and morbidity in patients with left ventricular dysfunction after myocardial infarction: results of the Survival and Ventricular Enlargement Trial. N Engl J Med 1992;327:669–77.
8. Hjalmarson A, Olsson G. Myocardial infarction. Effects of β-blockade. Circulation 1991; 84: 101–7.
9. The CAPRICORN investigators. Effect of carvedilol on outcome after myocardial infarction in patients with left-ventricular dysfunction: the CAPRICORN randomised trial. Lancet 2001;357:1385–90.
10. Prabhu SD, Chandrasekar B, Myrray DR, et al. β-Adrenergic blockade in developing heart failure: effects on myocardial inflammatory cytokines, nitric oxide, and remodeling. Circulation 2000; 101: 2103–9.
11. Beanlands RSB, Nahmias C, Gordon E, et al. The effects of β 1 -blockade on oxidative metabolism and the metabolic cost of ventricular work in patients with left ventricular dysfunction. Circulation 2000; 102: 2070–5.
12. Belichard P, Savard P, Cardial R, et al. Markedly different effects on ventricular remodeling result in decrease in inducibility of ventricular arrythmias. J Am Coll Cardiol 1994; 23: 505–13.
13. Hu K, Gaudron P, Ertl G. Long-term effects of beta-adrenergic blocking agent treatment on hemodynamic function and left ventricular remodeling in rats with experimental myocardial infarction: importance of timing of treatment and infarct size. J Am Coll Cardiol 1998; 31: 692–700.
14. Feuerestein GZ, Bril A, Ruffolo Jr. RR Protective effects of carvedilol in the myocardium. Am J Cardiol 1997; 80: 41L–45L.
15. Australia-New Zealand Heart Failure Research Collaboration Group. Randomized placebo-controlled trial of carvedilol in patients with congestive heart failure due to ischemic heart disease. Lancet 1997;349:375–80.
16. Packer M, Bristow MR, Cohn JN, et al. The U.S. Carvedilol Heart Failure Study Group. The effect of carvedilol on morbidity and mortality in patients with chronic heart failure. N Engl J Med 1996; 334: 1349–55.
17. Khaper N, Singal PK. Effect of afterload reducing drugs on the pathogenesis of antioxidant changes and congestive heart failure in rat. J Am Coll Cardiol 1997; 29: 856–61.
18. Stephens NG, Parsons A, Schofield PM, et al. Randomized controlled trial of vitamin E in patients with coronary disease: Cambridge Heart Antioxidant Study (CHAOS). Lancet 1996; 347: 781–86.
19. Hill MF, Singal PK. Antioxidant and oxidative stress changes during heart failure subsequent to myocardial infarction in rat. Am J Pathol 1996; 148: 291–300.
20. Ferrari R, Ceconi C, Curello S, et al. Oxygen-mediated myocardial damage during ischemia and reperfusion: role of the cellular defenses against oxygen toxicity. J Mol Cell Cardiol 1985; 17: 837–45.
21. Flesch M, Maack C, Cremers B, et al. The effect of β-blockers on free radical-induced cardiac contractile dysfunction. Circulation 1999; 100: 346–53.
22. Yamauchi-Takihara K, Ihara Y, Ogata A, et al. Hypoxic stress induces cardiac myocyte-derived interleukin-6. Circulation 1995; 91: 1520–4.
23. Liang C-S, Rounds NK, Dong E, et al. Alterations by norepinephrine of cardiac sympathetic nerve terminal function and myocardial β-adrenergic receptor sensitivity in the ferret. Circulation 2000; 102: 96–103.
24. Kinugawa S, Tsutsui H, Hayashidani S, et al. Treatment with dimethylthiourea prevents left ventricular remodeling and failure after experimental myocardial infarction in mice. Circ Res 2000; 87: 392–8.
25. Krown KA, Page MT, Nguyen C, et al. Tumor necrosis factor alpha-induced apoptosis in cardiac myocytes: involvement of sphingolipid signaling cascade in cardiac death. J Clin Invest 1996; 98: 2854–65.
26. Ferrari R, Agnoletti L, Comini L, et al. Oxidative stress during myocardial ischemia and heart failure. Eur Heart J 1999; 19 (suppl B): B2–B11.
27. Wei S, Chow LTC, Sanderson JE. Effect of carvedilol in comparison with metoprolol on myocardial collagen postinfarction. J Am Coll Cardiol 2000; 36: 276–81.
28. Nguyen QT, Cernacek P, Calderone A, et al. Endothelin A receptor blockade causes adverse left ventricular remodeling but improves pulmonary artery pressure after infarction in the rat. Circulation 1998; 98: 2323–30.
29. Robert R, Ruffolo Jr, RR Duane A, et al. Carvedilol (Kredex®): a novel multiple action cardiovascular agent. Drugs Today 1991; 27: 465–92.
30. Li KH, Rouleau J-L, Andries LJ, et al. Effect of dysfunctional vascular endothelium on myocardial performance in isolated papillary muscles. Circ Res 1993; 72: 768–77.
31. Nguyen T, El Salibi E, Rouleau JL. Postinfarction survival and inducibility of ventricular arrhythmias in spontaneously hypertensive rats: effects of ramipril and hydralazine. Circulation 1998; 98: 2074–80.
32. Tsoporis JN, Marks A, Kahn HJ, et al. S100β inhibits α 1 -adrenergic induction of hypertrophic phenotype in cardiac myocyte. J Biol Chem 1997; 272: 31915–21.
33. Zirin RA, Condra JH, Dixon RAF, et al. Molecular cloning and characterization of DNA sequences encoding rat and human atrial natriuretic factors. Proc Natl Acad Sci U S A 1984; 81: 6325–9.
34. Palmer JN, Hartogenesis WE, Patten M, et al. Interleukin-1β induces cardiac myocyte growth but inhibits cardiac fibroblast proliferation in culture. J Clin Invest 1995; 95: 2555–64.
35. Irwin MW, Mak S, Mann DL, et al. Tissue expression and immunolocalization of tumor necrosis factor-alpha in postinfarction dysfunctional myocardium. Circulation 1999; 99: 1492–8.
36. Luo XP, Yazdanpanah M, Bhooi N, et al. Determination of aldehydes and other lipid peroxidation products in biologic samples by gas chromatography mass spectrometry. Anal Biochem 1995; 228: 294–8.
37. Qi X, Li K, Rouleau JL. Endocardial endothelium and myocardial performance in rats: effects of changing extracellular calcium and phenylephrine. J Mol Cell Cardiol 1996; 28: 859–69.
38. Yue P, Massie, BM, Simpson PC, et al. Cytokines expression increases in nonmyocytes from rats with postinfarction heart failure. Am J Physiol 1998;275(Heart Circ Physiol 44):H250–8.
39. Tarazi RC, Sens S. Catecholamines and cardiac hypertrophy. In: Herzey KC, Caldwell ADS, eds. Catecholamines and the heart. London: Academic Press and Royal Society of Medicine, 1979: 47–57.
40. Grossman W, Jones D, McLaurin LP. Wall stress and pattern of hypertrophy in the human left ventricle. J Clin Invest 1975; 56: 56–64.
41. Vogt M, Strawer BE. Response of hypertensive left ventricular hypertrophy and coronary microvascular disease to calcium antagonists. Am J Cardiol 1995; 76: 24D–30D.
42. Veinot JP, Gattinger DA, Fliss H. Early apoptosis in human myocardial infarction. Hum Pathol 1997; 28: 485–92.
43. Sabbah HN, Sharov VG, Gupta RC, et al. Chronic therapy with metoprolol attenuates cardiomyocyte apoptosis in dogs with heart failure. J Am Coll Cardiol 2000; 36: 1698–705.
44. Feuerstein G, Yue TL, Ma X, et al. Novel mechanism in the treatment of heart failure: inhibition of oxygen-radical and apoptosis by carvedilol. Prog Cardiovasc Dis 1998; 41 (suppl 1): 17–24.
45. Orenstein TL, Parker TG, Butany JW, et al. Favorable left ventricular remodeling following large myocardial infarction by exercise training: effect on ventricular morphology and gene expression. J Clin Invest 1995; 96: 858–66.
46. Alpert NR, Mulieri LA, Hasenfuss G, et al. Optimization of myocardial function [review]. Basic Res Cardiol 1993; 88 (suppl 2): 29–41.
47. Swedberg K, Held P, Kjekshus J, et al. Effects of the early administration of enalapril on mortality in patients with myocardial infarction: results of the Cooperative New Scandinavian Enala Study II (CONSENSUS II). N Engl J Med 1992; 327: 678–84.
48. de Tombe PP, ter Keurs HE. Force and velocity of sarcomere shortening in trabeculae from rat heart: effect of temperature. Circ Res 1990; 66: 1239–54.

β-Blocker; Carvedilol; Cytokines; Postinfarction; Remodeling

© 2002 Lippincott Williams & Wilkins, Inc.