This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kajstura, J. Articles by Anversa, P. Search for Related Content PubMed PubMed Citation Articles by Kajstura, J. Articles by Anversa, P. Pubmed/NCBI databases Medline Plus Health Information Cardiomyopathy

(Circulation. 1995;92:2306-2317.)
© 1995 American Heart Association, Inc.

Articles

The Cellular Basis of Pacing-Induced Dilated Cardiomyopathy

Myocyte Cell Loss and Myocyte Cellular Reactive Hypertrophy

Jan Kajstura, PhD; Xun Zhang, MD; Yu Liu, MD; Ervin Szoke, MD; Wei Cheng, MD; Giorgio Olivetti, MD; Thomas H. Hintze, PhD; Piero Anversa, MD

From the Departments of Medicine and Physiology, New York Medical College, Valhalla, NY.

Correspondence to Piero Anversa, MD, Department of Medicine, Vosburgh Pavilion, Room 302, New York Medical College, Valhalla, NY 10595.

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Background Rapid ventricular pacing leads to a cardiac myopathy consisting of an increase in chamber dimension, mural thinning, elevation in ventricular wall stress, and congestive heart failure, mimicking dilated cardiomyopathy in humans. However, contrasting results have been obtained concerning the mechanisms of ventricular dilation and the existence of myocardial hypertrophy. Moreover, questions have been raised regarding the occurrence of myocardial damage and cell loss in the development of the experimental myopathy.

Methods and Results The functional and structural characteristics of the heart were studied in conscious dogs subjected to left ventricular pacing at 210 beats per minute for 3 weeks and 240 beats per minute for an additional week. At the time the animals were killed, measurements of myocardial structural integrity and myocyte shape, size, and number were determined by morphometric analysis of the myocardium in situ and enzymatically dissociated cells. The experimental protocol used was associated with overt cardiac failure documented by an increase in left ventricular end-diastolic pressure and a decrease in left ventricular systolic pressure and +dP/dt in combination with tachycardia, ascites, and pulmonary congestion. Although cardiac weights were not altered, cavitary diameter was increased and wall thickness was decreased from the base to the apex of the heart. Multiple foci of replacement fibrosis, comprising 6% of the myocardium, were detected across the left ventricular wall. Measurements of myocyte size and number documented a 39% loss of cells in the entire ventricle and a 61% increase in volume of the remaining viable myocytes. Myocyte hypertrophy was characterized by a 33% increase in cell length and a 23% increase in transverse area, resulting in a 23% increase in the cell length–to–cell diameter ratio. Pacing did not alter the relative proportion of mononucleated, binucleated, and multinucleated myocytes in the myocardium.

Conclusions Myocyte cell loss and myocyte reactive hypertrophy are the major components of ventricular remodeling in pacing-induced dilated cardiomyopathy.

Key Words: ventricles • myocytes • myocardium • pacing

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
Dilated cardiomyopathy defines a myocardial disease characterized by left ventricular or biventricular dilation, cardiac hypertrophy, and severely depressed myocardial performance.¹ In its late stage, tissue damage with collagen accumulation in the ventricular wall is present, but the magnitude of myocardial scarring contributes only in part to the expansion in cavitary volume.^{2 3 4 5 6} Myocyte elongation, through the in-series addition of newly formed sarcomeres, constitutes the major factor responsible for the increase in chamber size.⁷ Myocyte diameter also increases, although the lateral expansion of myocytes is modest and inadequate to preserve the wall thickness–to–chamber diameter ratio within normal values.⁷ As a consequence, decompensated eccentric hypertrophy characterizes the unfavorable evolution of idiopathic dilated cardiomyopathy in humans.^{7 8} Similarly, rapid ventricular pacing in animals leads to a cardiac myopathy consisting of an increase in chamber dimension, mural thinning, elevation in ventricular wall stress, alterations in coronary blood flow, and congestive heart failure,^{9 10 11 12 13 14} mimicking the human disease.^{1 7} However, contrasting results have been obtained concerning the mechanisms of ventricular dilation^{11 14} and the existence of myocardial hypertrophy.^{11 14 15 16} Measurements of myocyte length in paced pigs have indicated a consistent increase in this parameter, whereas myocyte width has been found to be decreased.^{11 17} Moreover, questions have been raised regarding the occurrence of myocardial damage and cell loss in the development of experimental myopathy.^{11 14 16 18} Importantly, measurements of myocyte size and number have not been obtained in this model, and scattered myocyte loss and collagen accumulation have been shown to play an essential role in the onset and progression of ventricular dysfunction in ischemic heart disease,^{19 20} pressure-overload hypertrophy,²¹ aging,²² and a combination of these multiple variables in humans²³ and animals.²⁴ Myocyte loss may occur in a diffuse, scattered manner, becoming apparent only by quantitative analysis of the myocardium.^{19 22} This is not an unlikely possibility because a significant component of pacing-induced heart failure involves defects in endothelium-mediated responses of the coronary circulation.²⁵ Such an impairment in the vasodilatory capacity of intermediate-sized arteries and arterioles may generate focal areas of ischemia across the ventricular wall, resulting in necrotic myocyte cell death. A continued loss of myocytes can be expected to produce a greater workload on the remaining myocytes,^{19 22} which, coupled with a increase in systolic and diastolic wall stress,^{10 14} may serve as a chronic mechanical stimulus for cellular growth. Therefore, the effects of rapid ventricular pacing on tissue injury, myocyte size, shape, and number were analyzed in dogs with overt signs of congestive heart failure to determine whether myocyte cell loss and cellular enlargement were relevant variables of wall restructuring in this type of cardiac myopathy.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
Animals
Mongrel dogs weighing 18 to 20 kg were instrumented with catheters, probes, and a corkscrew electrode in the left ventricle attached to a portable external pacemaker (Pace Medical EV3434), as described previously.^{25 26} Briefly, each dog was sedated with acepromazine (1 mg/kg SC) and then anesthetized with sodium pentobarbital (25 mg/kg IV). A Tygon catheter (Cardiovascular Instruments) was placed in the descending thoracic aorta, and a solid-state manometer (Konigsberg P6.5) was placed in the left ventricle through the apex, with the use of sterile surgical techniques. The dogs were allowed to recover from surgery for 10 to 14 days. Initial experiments were conducted when they were afebrile and had been trained to lie quietly without restraint on the laboratory table. The heart then was paced at 210 beats per minute for 3 weeks and at 240 beats per minute for an additional week. The heart was initially paced at 210 beats per minute to achieve a mild form of ventricular dysfunction and subsequently at 240 beats per minute to obtain physiological parameters and clinical symptoms of congestive heart failure (for review, see Reference 27). The control group was similarly instrumented, but the heart was not paced. Sixteen sham-operated control dogs and 14 paced dogs were included in this study. The protocols were approved by the Institutional Animal Care and Use Committee of New York Medical College and conform to the Guiding Principles for the Use and Care of Laboratory Animals of the American Physiological Society and the National Institutes of Health.

For the objective of this study, it is important to indicate that the electrode was inserted in the midportion of the left ventricular free wall near the apex and its location confirmed at surgery. At the dissection of the heart, the wire connecting the pacing lead was cut, but the electrode was not removed. This reference point allowed sampling of tissue for histological analysis at a distance of at least 3 cm from the electrode. Moreover, myocardial sections obtained from the base of the heart were significantly more distant, and the interventricular septum was completely separated from the electrode.

Ventricular Function
Hemodynamic studies were performed with the dogs in a conscious state with the pacemaker turned off. Once the steady state was reached, the hemodynamic measurements were made. Systemic arterial pressure, left ventricular systolic pressure, left ventricular end-diastolic pressure, and left ventricular dP/dt were obtained.²⁶ Specifically, before these determinations, the pacemaker was turned off for nearly 2 hours. This is the time required for coronary blood flow and especially heart rate to reach a constant level.

Perfusion Fixation
At the completion of the hemodynamic determinations, hearts were perfusion-fixed in situ. Dogs were anesthetized with sodium pentobarbital (25 mg/kg IV) and mechanically ventilated. The thorax was opened in the left fourth intercostal space. A catheter was placed in the left fourth intercostal space. A catheter was placed in the left ventricle through a stab wound. Heparin (3000 units, Lyphomed) was injected intravenously, and the heart was arrested with a bolus injection of saturated potassium chloride. The aorta was cannulated for retrograde perfusion, and the right atrial appendage was cut to allow outflow of perfusate. Perfusion pressure was adjusted to the mean arterial pressure measured in vivo. The left ventricular chamber was filled with fixative and kept at a pressure equal to the in vivo measured end-diastolic pressure throughout the fixation procedure.²⁸ After perfusion with pH 7.2 phosphate buffer for 3 minutes, the coronary vasculature was perfused for 10 minutes with a solution containing 2% paraformaldehyde and 2.5% glutaraldehyde. Subsequently, the heart was excised and the weights of the left ventricle, interventricular septum, and the right ventricle were recorded. The separate measurement of left ventricular free wall and septum weights were obtained following the estimations of ventricular dimensions listed below. This part of the study included 7 controls and 7 paced hearts.

Ventricular Dimensions
The major intracavitary axis of the left ventricle was measured by inserting a probe from the base to the apex parallel to the longitudinal diameter. Care was taken to avoid inclusion of muscle at the apical region in the estimation of this parameter. Subsequently, each left ventricle was serial-sectioned into thick rings perpendicular to the longitudinal axis to obtain six equally spaced parallel planes from the base to the apex in which the average wall thickness of the free wall and septum and chamber luminal diameter were measured. These determinations could not be obtained at the two extremes of the left ventricle, the apex, and base. The trabeculae carnae and papillary muscles attached to the wall were not included in the assessment of wall thickness. Fifteen measurements each in the anterior and posterior aspects of the left ventricular free wall and interventricular septum were collected in each section and their values averaged for each of the three regions. The minimal and maximal diameters of the ventricular chamber at each of the six sampled levels were determined, and their geometric mean was computed.²⁹ Moreover, the measurements of wall thickness, chamber radius, and left ventricular end-diastolic pressure were used to compute midwall circumferential diastolic wall stress at each of the six sites examined from base to apex.

Light Microscopic Morphometry
Tissue samples were obtained from the anterior aspect of the left ventricular free wall and interventricular septum. Four tissue fragments were collected from section 1 near the base, section 4 in the middle, and section 6 adjacent to the apex, for both the free wall and septum. Thus, 12 tissue blocks for the ventricle and 12 for the septum were sampled. These specimens were embedded in paraffin; histological sections were prepared, and these sections were stained with hematoxylin and eosin and trichrome for morphometric analysis. In these sections, which contained the entire thickness of wall and septum, 20 consecutive fields, each from the endomyocardium, midmyocardium, and epimyocardium of the wall and the left, intermediate, and right side of the septum, were examined at a calibrated magnification of x400 with a reticle containing 42 sampling points (105844, Wild Heerbrugg Instruments Inc). This reticle defined an uncompressed tissue area of 62 500 µm², which was used to determine the number of lesions represented by foci of damage per unit area of myocardium. Moreover, the fraction of points lying over these foci was measured to compute the volume fraction of lesions in the myocardium. These sections also were used to measure the volume fraction of myocytes and interstitium in regions of the wall and septum in which tissue injury was not present. This analysis was performed by examining 10 fields in each area of the myocardium.

Myocyte Isolation
In a separate group of control (n=9) and paced hearts (n=7), myocytes were enzymatically dissociated for the evaluation of cell size and shape. This analysis was restricted to the left ventricular free wall. At the end of the hemodynamic measurements, hearts were removed, a portion of the anterior aspect of the left ventricle was dissected free, and a large branch of the left anterior descending artery was identified and cannulated for perfusion with collagenase. The solutions were supplements of modified commercial MEM Eagle Joklik. HEPES-MEM contained (in mmol/L) NaCl 117, KCl 5.7, NaHCO₃ 4.4, KH₂PO₄ 1.5, MgCl₂ 17, HEPES 21.1, and glucose 11.7, with amino acids and vitamins, 2 mmol/L L-glutamine, and 21 mU/mL insulin; pH was adjusted to 7.2 with NaOH. Osmolality of this solution is 292 mOsm. Resuspension medium was HEPES-MEM supplemented with 0.5% bovine serum albumin, 0.3 mmol/L calcium chloride, and 10 mmol/L taurine, adjusted to 292 mOsm. The cell isolation procedure consisted of three main steps. (1) Low calcium perfusion: Blood washout and collagenase (selected type II, Worthington Biochemical Corp) perfusion of the myocardium was carried out at 32°C with HEPES-MEM gassed with 85% O₂, 15% N₂. (2) Mechanical tissue dissociation: After removing the myocardium from the cannula, the endomyocardium and epimyocardium were separated and minced. Collagenase-perfused tissue was subsequently shaken in resuspension medium containing creatine, collagenase, and 0.3 mmol/L calcium chloride. Supernatant cell suspensions were washed and resuspended in resuspension medium. (3) Separation of intact cells: Intact cells were enriched by centrifugation and discarding the supernatant. This procedure was repeated four to five times in each preparation in order to remove nonmyocyte cells, cell debris, and residual collagenase. Each centrifugation was performed at 30g for 3 minutes. Rectangular trypan blue, excluding cells, constituted around 75% to 85% of all myocytes. The average number of myocytes collected per 1 g of myocardium was 7 to 8x10⁶ and 4 to 5x10⁶ in control and paced hearts, respectively (Fig 1).

View larger version (93K): [in this window] [in a new window]   Figure 1. Photomicrograph of enzymatically dissociated left ventricular myocytes obtained from the anterior aspect of the left ventricular free wall of a paced heart with congestive heart failure.

Distribution of Nuclei and Myocytes
Intact cells were recovered from the pellet, washed, and fixed in 2.5% glutaraldehyde in phosphate buffer. Approximately 50 to 100x10³ myocytes were suspended in bis-benzimide H33258, 500 µg/mL, for 15 minutes at room temperature. Subsequently, myocytes were centrifuged at 100g for 5 minutes, the supernatant was discarded, and smears were made from these preparations. This was performed with a microscope working in epifluorescence mode, equipped with a set of excitation-emission filters for bis-benzimide H33258. A random sampling of 1000 myocytes from the left ventricle in each heart was used to determine the relative frequency of mononucleated, binucleated, and multinucleated cells.

Structural Properties of Myocytes
Measurements of myocyte geometric dimensions were accomplished through the aid of a computerized image analysis system (Morrell Instruments): 200 myocytes from each left ventricle were measured to obtain length, width, and area. Moreover, cell volume was derived from these geometric parameters. The distribution of myocytes isolated from the left ventricle for each of the two animal groups was divided according the number of cells in an established range of lengths. The range of lengths used in the present study was 10 µm from 70 to 320 µm. Histogram buckets then were established with a size of 10 µm, and frequency distribution histograms were constructed by plotting number of cells on the ordinate and cell length range on the abscissa. A similar approach was followed for the analysis of the distribution of cell cross sections and volumes. Moreover, average sarcomere length was calculated from the linear distance encompassing between 25 and 50 contiguous sarcomeres in 100 cells per ventricle measured at a final magnification of x1000. These determinations were obtained in binucleated myocytes. An additional sampling consisting of 20 each mononucleated, trinucleated, and tetranucleated myocytes was collected in each heart.

For the estimation of myocyte volume, the following procedure was used. Isolated cells when placed in physiological medium assume a cross-sectional area, which resembles a flattened ellipse. The ratio of the minor axis (b) to the major axis (a) of the ellipse was 0.384±0.065 in myocytes from control dogs and 0.413±0.088 in myocytes from paced animals. The ratio between the minor and major axis was determined by confocal microscopy (see below). Cell volume (V_C) was calculated assuming an elliptical cross section with a major axis that was equivalent to cell width and a minor axis that was computed from the measured ratios. Cell length (L) was measured directly:

(1)

Confocal Microscopic Measurements of Myocyte Cell Volume
The quantitative analysis discussed above to obtain measurements of the average volume of mononucleated, V(c)_n, binucleated, V(c)_2n, trinucleated, V(c)_3n, and multinucleated, V(c)_mn, myocytes in the ventricular myocardium was complemented with a new procedure. Isolated myocytes were stained with fluorescein isothiocyanate 1 µg/mL for 30 minutes at room temperature to visualize the cell cytoplasm and with propidium iodide to label the nuclei. Subsequently, measurements of cell area for mononucleated, (c)_n, binucleated, (c)_2n, trinucleated, (c)_3n, and multinucleated, (c)_mn, myocytes were collected by confocal microscopy. Importantly, with the same system, it was possible to measure the thickness, (c)_n, (c)_2n, (c)_3n, and (c)_mn, of these different cell populations by optical sectioning of each cell in the Z plane. Therefore, the average volume of each cell type, V(c)_n, V(c)_2n, V(c)_3n, and V(c)_mn, was computed from

(2)

The confocal measurements of cell volume were restricted to 5 mononucleated, 16 binucleated, 7 trinucleated, and 5 tetranucleated myocytes in control left ventricles. Corresponding values in paced hearts were 5 mononucleated, 19 binucleated, 4 trinucleated, and 4 tetranucleated. The cells examined were randomly sampled by measuring only myocytes vertically oriented in the microscopic field. The proportion between the width and thickness of myocytes was used to obtain the average cell volume in the larger population of cells measured in the preceding section (see above).

Number of Myocytes
The aggregate number of myocytes in the left ventricular free wall was obtained by combining measurements obtained by the in situ analysis of the myocardium with the estimations of myocyte cell volume, collected from enzymatically dissociated cells. Specifically, the total volume of the left ventricular myocardium, V_T, was first determined by dividing its weight by the specific gravity of muscle tissue, 1.06 g/mL (Reference 30). Subsequently, the absolute volume of viable tissue, V, was assessed from the product of the ventricular volume and the volume percent of myocardium not affected by replacement fibrosis, V_v:

(3)

The total volume of myocytes in the left ventricular free wall, V(c)_T, then was calculated from V and the volume fraction of myocytes in the viable tissue, V(c)_V:

(4)

From the morphometric measurement of the volume fraction of myocytes in the myocardium, V(c)_V, and the proportion of mononucleated, F(c)_n, binucleated, F(c)_2n, trinucleated, F(c)_3n, and multinucleated, F(c)_mn, cells determined in enzymatically dissociated myocytes, the volume percent of mononucleated, V(c)_Vn, binucleated, V(c)_V2n, trinucleated, V(c)_V3n, and multinucleated, V(c)_Vmn, cells in the tissue was obtained:

(5)

This information combined with the quantitative evaluation of the absolute volume of myocytes in the ventricle, V(c)_T (see Equation 4), allowed the estimation of the aggregate volume of mononucleated, V(c)_Tn, binucleated, V(c)_T2n, trinucleated, V(c)_T3n, and multinucleated, V(c)_Tmm, cells:

(6)

Finally, the number of mononucleated, N(c)_n, binucleated, N(c)_2n, trinucleated, N(c)_3n, and multinucleated, N(c)_mn, cells in the ventricle was computed.

(7)

Data Collection and Analysis
All tissue samples were coded, and the code was broken at the end of the experiment. Results are presented as mean±SD computed from the average measurements obtained from each dog. Statistical significance for comparisons between two measurements was determined with the unpaired two-tailed Student's t test. Statistical significance for comparisons among measurements within the wall of each ventricle was determined with one-way ANOVA. Statistical significance in multiple comparisons among independent groups of data in which ANOVA and the F test indicated the presence of significant differences was determined by the Bonferroni method.³¹ Values of P<.05 were considered significant. Because measurements presented were not obtained in all animals, n values for each parameter determined are listed in the text or the legend of each figure.

	Results

Top Abstract Introduction Methods Results Discussion References

 
Ventricular Function
Fig 2 illustrates that ventricular pacing at the regimen performed here resulted in a significant impairment of cardiac performance. Heart rate was increased 41% (controls, 85±18 beats per minute; paced, 120±17 beats per minute; P<.0001), and mean arterial pressure was reduced 15% (P<.0001). Moreover, left ventricular systolic pressure was decreased 24% (P<.0001), whereas left ventricular end-diastolic pressure was increased 295% (P<.0001). Left ventricular positive dP/dt was 46% (P<.0001) lower in the paced group. These alterations in physiological parameters of ventricular dynamics were all obtained in dogs in a conscious state and with the pacemaker turned off. Of relevance, animals subjected to ventricular pacing gained weight during the interval examined and displayed clinical symptoms and signs of severe pump failure. Tachycardia, tachypnea, ascites, pulmonary congestion, and pleural effusion was present in these animals. In summary, ventricular pacing for 4 weeks resulted in congestive heart failure.

View larger version (39K): [in this window] [in a new window]   Figure 2. Bar graphs show effects of rapid ventricular pacing on functional properties of the heart. Results are presented as mean±SD. *Value is statistically significantly different, P<.05. Controls, open bars, n=16; paced hearts, hatched bars, n=14. LVSP indicates left ventricular systolic pressure; LVEDP, left ventricular end-diastolic pressure; and LV +dP/dt, left ventricular positive dP/dt.

Cardiac Anatomy
The weights of the left ventricular free wall, interventricular septum, and right ventricle of control hearts were 81±12 g, 41±6 g, and 48±7 g. Corresponding values in paced hearts were 83±21 g, 42±14 g, and 52±9 g. None of the differences found between the two groups of animals were statistically significant. However, the lack of changes at the organ weight level after pacing did not exclude modifications in wall thickness, chamber diameter, and longitudinal axis, which characterize ventricular size and shape. With the exception of the estimation of the longitudinal axis, which constitutes a single measurement, wall thickness and chamber diameter of the left ventricle were obtained in six subsequent rings from base to apex. The thickness of the interventricular septum was analyzed in a similar manner. The apical and basal portions were not included in this analysis because of the impossibility of obtaining these parameters at the two ends of the heart. Fig 3 illustrates that the thickness of the anterior aspect of the left ventricular free wall decreased 21% (P<.01), 18% (P<.006), 18% (P<.006), 14% (NS), 16% (NS), and 20% (P<.002) from base to apex. Corresponding changes in the posterior portion of the ventricle were -16% (NS), -16% (P<.02), -17% (P<.02), -13% (NS), -20% (P<.03), and -20% (P<.005). The thickness of the interventricular septum also consistently decreased at all levels. When the individual values were combined to compute an average measurement of the anterior and posterior regions of the ventricle and septum, wall thickness was decreased 18% (P<.003), 17% (P<.005), and 16% (P<.001), respectively.

View larger version (67K): [in this window] [in a new window]   Figure 3. Bar graphs show effects of rapid ventricular pacing on the thickness of the anterior (A) and posterior (B) aspects of the left ventricular free wall and interventricular septum (C). Changes in left ventricular chamber diameter are shown in D. Results are presented as mean±SD. *Value is statistically significantly different, P<.05. Controls, open bars, n=7; paced hearts, hatched bars, n=7.

The modifications in chamber diameter as a result of ventricular pacing are shown last in Fig 3. This parameter increased mostly in the ring adjacent to the base, 35% (P<.0001), whereas the changes in the two rings near the apex were not statistically significant. The expansions at levels 2, 3, and 4 were 27% (P<.0002), 25% (P<.0001), and 21% (P<.004). As an average, chamber diameter increased 23% (P<.02) in paced hearts. These alterations provoked an 85% (P<.0001), 63% (P<.0001), 64% (P<.0001), 55% (P<.0001), 50% (P<.02), and 46% (P<.05) increase in chamber diameter–to–wall thickness ratio from base to apex (data not shown). In addition, the major longitudinal axis of the left ventricle increased 7% (P<.04), from 59±3.1 mm to 63±2.7 mm. The changes in transverse and longitudinal chamber diameters implied a 62% increase in cavitary volume. The combination of these anatomic changes with the increase in left ventricular end-diastolic pressure generated a significant elevation in diastolic wall stress at all levels, from base to apex (Fig 4). In summary, ventricular pacing for 4 weeks resulted in chamber dilation, wall thinning, and a marked increase in diastolic wall stress.

View larger version (33K): [in this window] [in a new window]   Figure 4. Bar graph shows effects of rapid ventricular pacing on the distribution of midwall circumferential diastolic wall stress. Results are presented as mean±SD. *Value is statistically significantly different, P<.05. Controls, open bars, n=7; paced hearts, hatched bars, n=7.

Myocardial Damage
The observations above indicating that major changes occurred in ventricular anatomy with pacing, in the absence of alterations in gross cardiac weights, suggested that myocardial injury and myocyte reactive hypertrophy were present. Alternatively, modifications in myocyte shape without cellular hypertrophy could have occurred. Fig 5A illustrates that multiple foci of replacement fibrosis due to discrete losses of myocytes were found across the ventricular wall in all paced dogs. The interventricular septum also was affected, but tissue damage was mostly restricted to the left portion of the septum. The sites of injury were characterized by areas of scarring, with little or no vessel profiles and inflammatory cells, or by more recent lesions with fibroblast and capillary proliferation, leukocytes, and mononuclear infiltrates (Fig 5B). Foci of acute myocytolytic necrosis were also detected (Fig 5C).

View larger version (144K): [in this window] [in a new window]   Figure 5. Photomicrographs show effects of rapid ventricular pacing on myocardial structural integrity. Foci of replacement fibrosis are illustrated in A, whereas a more acute lesion with multiple inflammatory cells is depicted in B. Myocytolytic necrosis is shown in C. Hematoxylin and eosin staining: A, x180; B, x650; C, x950.

Fig 6 illustrates the volume percent (Fig 6A and 6B) and numerical density (Fig 6C and 6D) of foci of replacement fibrosis in the left ventricular free wall. Although these parameters were different from control values, regional variations in the three principal layers of the wall were not demonstrable on a statistical basis (Fig 6A and 6C). The endomyocardium exhibited consistently higher values than the midmyocardium and the epimyocardium, but these differences did not reach statistical significance. An identical evaluation performed in the interventricular septum is shown in Fig 6B and 6D. The left portion exhibited more extensive damage than the intermediate and right side of the septum. Measurements of the volume composition of the myocardium in regions not affected by replacement fibrosis showed that the percentage of myocytes was 80±2% in controls and 79±2% in paced hearts. Corresponding values for the nonmyocyte compartment were 20±2% and 21±2%. In summary, ventricular pacing for 4 weeks resulted in chronic damage of the myocardium of the left ventricular free wall and interventricular septum.

View larger version (33K): [in this window] [in a new window]   Figure 6. Bar graphs show effects of rapid ventricular pacing on the volume percent (A and B) and numerical density (C and D) of foci of myocardial damage represented by replacement fibrosis in the endomyocardium (EN), midmyocardium (M), epimyocardium (E), and wall (W) of the left ventricle. The distribution of the same parameters in the left side (LS), middle portion (MP), and right side (RS) of the septum and in the entire septum (S) are also shown. Results are presented as mean±SD. *Value is statistically significantly different, P<.05. Controls, open bars, n=7; paced hearts, hatched bars, n=7.

Proportions of Myocyte Populations in the Myocardium
The analysis of the distribution of mononucleated, binucleated, trinucleated, and multinucleated myocytes in the left ventricular free wall of control and paced dogs is shown in Fig 7. Binucleated myocytes comprised the majority of cells in both groups of animals, and mononucleated cells were the second most frequent cell type. Significantly smaller components were represented by trinucleated and multinucleated myocytes. Ventricular pacing was associated with no statistically significant changes in any group of cells. Sarcomere length remained constant in all cells, averaging 1.88±0.10 µm (n=7) and 1.89±0.09 µm (n=7) in myocytes from control and paced hearts, respectively. In summary, ventricular pacing for 4 weeks did not alter the proportion of the different myocyte populations in the ventricular myocardium.

View larger version (20K): [in this window] [in a new window]   Figure 7. Bar graph shows effects of rapid ventricular pacing on the myocyte populations in left ventricle. Results are presented as mean±SD. Controls, open bars, n=7; paced hearts, hatched bars, n=7. 1N indicates mononucleated myocytes; 2N, binucleated myocytes; 3N, trinucleated myocytes; and 4N, tetranucleated myocytes.

Myocyte Size and Shape
A range of muscle cell lengths was observed in enzymatically dissociated left ventricular myocytes obtained from the anterior aspect of the left ventricular free wall in control and paced hearts. The interventricular septum was not included in this part of the study. Moreover, the description of myocyte parameters given below was restricted to binucleated myocytes of control and paced animals. An analysis of the distribution of myocyte lengths of control animals demonstrated that this parameter was evenly balanced around the mean value (Fig 8A). Moreover, the majority of the cells measured were close to the mean. After ventricular pacing, there was a shift to the right of this curve, which indicated that an increase in average cell length occurred in failing dogs (Fig 8B). Only a small fraction of cells was shorter than 120 µm in this group. On the other hand, this cell length comprised a significant fraction of myocytes from control hearts. Fig 8C and 8D shows the changes in the distribution of the cross-sectional area of myocytes after ventricular pacing. In comparison with control myocytes, myocyte transverse area in paced dogs was shifted to the right, revealing the presence of a population of cells wider than those seen in nonpaced animals. Cells from failing hearts documented a distribution of cell area that encompassed a greater range than that seen in controls. A similar shift to the right was detected in the distribution of myocyte cell volume with pacing, demonstrating that myocyte cellular hypertrophy occurred in this model (Fig 8E and 8F).

View larger version (37K): [in this window] [in a new window]   Figure 8. Bar graphs show effects of rapid ventricular pacing on the distribution of myocyte length, cross-sectional area, and volume. Controls, open bars, n=7; paced hearts, hatched bars, n=7.

Average length and cross-sectional area of binucleated myocytes in each heart were computed from these individual measurements to yield a mean value in each of the two groups of animals (Table). Pacing resulted in a 33% increase in length of binucleated myocytes. When this length change was examined in terms of number of sarcomeres, paced myocytes possessed 24 more sarcomeres than control cells (control, 72±5; paced, 96±4; 32%; P<.001). As a consequence of pacing, myocyte cross-sectional area increased 23%, and the phenomenon resulted in a 23% (P<.05) increase in the length-to-diameter ratio of myocytes. The changes in length and cross-sectional area of mononucleated, trinucleated, and tetranucleated myocytes in control and paced hearts are also listed in the Table. Cell length was increased in these myocytes, whereas the transverse area expanded in mononucleated cells and decreased in tetranucleated myocytes. These changes implied a variation in cell configuration with pacing.

View this table: [in this window] [in a new window]   Table 1. Effects of Ventricular Pacing on Myocyte Dimensional Properties

Fig 9 illustrates that ventricular pacing produced a 61% (P<.004) increase in volume of binucleated myocytes that represented the larger fraction of the myocyte population. Moreover, mononucleated myocytes that constituted the second most frequent cell type enlarged by 158% (P<.0001). Conversely, trinucleated and tetranucleated myocytes did not exhibit significant changes in cell volume. In summary, ventricular pacing for 4 weeks resulted in myocyte cellular hypertrophy.

View larger version (38K): [in this window] [in a new window]   Figure 9. Bar graph shows effects of rapid ventricular pacing on the average volume of mononucleated (1N), binucleated (2N), trinucleated (3N), and tetranucleated (4N) myocytes. Results are presented as mean±SD. *Value is statistically significantly different, P<.05. Controls, open bars, n=7; paced hearts, hatched bars, n=7.

Myocyte Dimensions by Confocal Microscopy
Fig 10 illustrates 16 optical sections, 1 µm apart, of a myocyte obtained from a paced dog. By this three-dimensional section reconstruction, myocyte cell volume was measured and, in this specific case, was found to be 40 418 µm³. It also should be apparent that with this approach, information was obtained concerning the relationship between cell width and cell thickness (Fig 11), which was used to compute cell cross sections and volumes illustrated in Fig 8E and 8F. Based on the analysis of randomly sampled myocytes, the average volumes of mononucleated (control, 16 200±3300 µm³, n=5; paced, 27 400±8000 µm³, n=5; 69%, P<.02), binucleated (control, 25 800±2200 µm³, n=16; paced, 38 000±3400 µm³; 47%, P<.0001), trinucleated (control, 47 500±7000 µm³, n=7; paced, 52 400±4800 µm³, n=4; 10%, NS), and tetranucleated (control, 55 000±7400 µm³, n=5; paced, 54 800±8700 µm³, n=4; 0%, NS) cells also were measured by confocal microscopy in control and paced dogs. These cellular parameters were comparable to the mean values of each cell population described in the preceding section. In summary, the hypertrophic growth of myocytes after 4 weeks of ventricular pacing was confirmed through three-dimensional optical section reconstruction by confocal microscopy.

View larger version (62K): [in this window] [in a new window]   Figure 10. Three-dimensional optical section reconstruction by confocal microscopy of a left ventricular myocyte from a paced heart. Magnification x300.

View larger version (43K): [in this window] [in a new window]   Figure 11. Confocal microscopic images of cross-sectional areas of two flattened myocytes on the surface of a microscopic slide. These were obtained by three-dimensional optical reconstruction on the Z plane of the cells. Magnification x1400.

Myocyte Number
Although the results described above documented that the experimental protocol was associated with tissue injury and myocyte hypertrophy, the question remained concerning the extent of myocyte loss that occurred in pacing-induced heart failure. The evaluation of the volume percent and numerical density of lesion profiles in the myocardium discussed in the preceding section did not allow a computation of the number of myocytes lost in the ventricle. This is because, as a consequence of myocytolytic necrosis, there are variable and unknown amounts of swelling in cells and interstitial space and a progressive lateral and radial shrinkage of the damaged area.⁸ Scar formation, the contraction of necrotic tissue with time, and reactive growth responses in the unaffected myocardium all represent dynamic processes that continuously change the proportion of viable and nonviable myocardium in the ventricle. However, the availability of left ventricular free wall weight, magnitude of myocardial damage, volume percent of myocytes in the myocardium, and average cell volume of the different myocyte populations allowed the estimation of the total number of each cell type in the ventricle.

This was accomplished by calculating first the fraction of tissue occupied by the various myocyte populations in the ventricle. The percentage of myocardium constituted by mononucleated myocytes was 1.94±0.82% and 2.29±0.17% in control and paced hearts, respectively. Corresponding values for binucleated myocytes were 74.50±1.61% and 71.12±2.47%. Trinucleated and tetranucleated cells comprised significantly smaller portions of the tissue. These two cell types involved 0.80±0.52% and 2.79±1.38% in controls and 1.95±0.86% and 3.88±2.23% in experimental animals. Only the differences between binucleated (P<.01) and trinucleated (P<.01) myocytes were statistically significant. Subsequently, the aggregate volume of myocytes represented by mononucleated (controls, 1.47±0.62 cm³; paced, 1.67±0.12 cm³, NS), binucleated (controls, 56.52±1.22 cm³; paced, 52.01±1.80 cm³; P<.0001), trinucleated (controls, 0.61±0.40 cm³, paced, 1.43±0.63 cm³; P<.05), and tetranucleated (controls, 2.12±1.01 cm³; paced, 2.84±1.63 cm³, NS) cells in the ventricle was obtained. This parameter, in combination with the mean cell volume of each myocyte population, allowed the calculation of the total number of each cell group in the heart. Fig 12 illustrates that the left ventricular free wall of control dogs contains 2.34x10⁹ myocytes, whereas this parameter was reduced to 1.42x10⁹ in paced animals, mostly because of a 47% and 40% reduction in mononucleated (P<.05) and binucleated cells (P<.0001). These changes resulted in a 39% (P<.0001) loss in the aggregate number of ventricular myocytes. In summary, ventricular pacing for 4 weeks resulted in a significant amount of myocyte cell loss in the myocardium.

View larger version (32K): [in this window] [in a new window]   Figure 12. Effects of rapid ventricular pacing on the aggregate number of mononucleated, binucleated, trinucleated, and tetranucleated myocytes. Results are presented as mean±SD. *Value is statistically significantly different, P<.05. Controls, open bars, n=7; paced hearts, hatched bars, n=7.

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
The results of the current study demonstrate that chronic severe tachycardia was associated with congestive heart failure and a marked elevation in diastolic wall stress from the base to the apex of the left ventricle. This abnormal hemodynamic state was accompanied by extensive ventricular remodeling consisting of chamber dilation and mural thinning, which affected the ventricle at all levels along the major longitudinal axis. Moreover, discrete areas of myocardial injury characterized by sites of acute myocytolytic necrosis and multiple foci of replacement fibrosis in various phases of healing were detected across the ventricular wall. The magnitude of myocyte loss was considerable and involved both mononucleated and binucleated cells. The myocyte cellular hypertrophic response was commensurate with the extent of myocyte loss, resulting in a preservation of cardiac weights. Importantly, myocyte enlargement included increases in myocyte length and diameter, but myocyte elongation was the prevailing form of reactive cell growth. Thus, the anatomic modifications coupled with ventricular pacing are the consequence of three structural events: myocyte cell loss, myocyte cellular hypertrophy, and an increase in cell length–to–diameter ratio.

Myocyte Cell Loss and Ventricular Function
Data in this investigation indicate that sustained tachycardia resulted in a 39% loss of myocytes in a period of 4 weeks, and this phenomenon was associated with left ventricular failure. The mural distribution of myocardial damage in the left ventricle provided direct evidence that clusters of myocytes were lost in the inner, middle, and outer layers of the wall. The volume percent and numerical density of lesion profiles were consistently higher in the endomyocardium than in the midmyocardium and epimyocardium, suggesting that myocyte cell death occurred prevalently in the subendocardial region of the ventricle. Important differences appear to exist between the effects of left ventricular pacing used here and those of right ventricular pacing.^{14 15} In the latter case, tissue injury and myocyte loss have not been observed, and the severe depression in cardiac pump function has been attributed to a number of factors, including impairment of subendocardial flow and vasodilator reserve mediated by the marked elevation in left ventricular end-diastolic pressure.¹³ In addition, structural arrangements of myofibrils in myocytes have been claimed as critical components of the reduction in force generating ability of the myocardium.¹⁴ Defects in the regulation and signaling pathway of ß-adrenergic receptors have been documented^{12 32} in combination with alterations in calcium handling³³ of the myopathic cardiac muscle. These cellular abnormalities have been linked to the altered diastolic function in this model¹⁶ and are considered to play a major role in the development of the decompensated dilated heart.^{13 32 33} However, tissue injury with collagen accumulation in the ventricular wall has been observed previously and proposed as a relevant mechanism of ventricular dysfunction and failure.¹⁸ Indices of myocyte cell loss have also been obtained in pigs after rapid ventricular pacing.¹¹ Although differences in the duration of pacing, location of the electrode, and program of stimulation exist in these studies, it is difficult to reconcile these various observations.

In recent years, several studies in humans^{7 20 22 23} and animals^{19 21 24 34 35 36 37} have demonstrated that myocyte cell loss is a consistent finding of cardiac failure of various etiology. Myocyte loss accompanies the evolution of different myopathies and appears to precede the impairment in ventricular pump function.^{34 38 39} Moderate losses of myocytes, scattered throughout the wall, have a profound impact on ventricular performance.¹⁹ Myocyte death, diffuse in nature, seems to have a greater influence on cardiac hemodynamics than an equivalent loss of myocytes in a segmental fashion.¹⁹ Experimental results^{40 41} and observations in humans^{42 43 44} have indicated that occlusion of a major coronary artery resulting in acute myocardial infarction leads to overt failure when the destruction in muscle mass involves 40% to 50% of the myocyte population of the left ventricle. Moreover, it has been shown that nonocclusive coronary artery constriction, coupled with a 20% diffuse myocyte death, is associated with alterations in cardiac hemodynamics that are not commonly seen with infarcts of comparable size.¹⁹ Similar results have been obtained after systemic hypertension.^{21 39} At present, there is no information concerning the minimal magnitude of cell loss required to depress cardiac performance in dogs when cell death occurs segmentally or in a scattered manner across the wall. However, the current findings strongly suggest that a 39% reduction in the aggregate number of myocytes is associated with the terminal phases of the cardiac myopathy. This may constitute a difference between the experimental myopathy examined here and that in humans.⁷ Ventricular weight is relatively unchanged in the dog model, whereas it is markedly increased in humans. Such a phenomenon may be dependent on the variability in the duration of the disease process, the magnitude of myocyte loss, the extent of reactive myocyte growth, or a combination of these factors.

Results in this investigation cannot establish whether the 6.37% of replacement fibrosis measured across the left ventricular wall accounted for the entire 39% loss of myocytes with rapid ventricular pacing. On the basis of observations made in other animal models^{8 19 21 38} and humans,^{7 22 39} this extent of myocardial damage has to be considered responsible for a large fraction of myocyte cell death but cannot be interpreted as the exclusive mechanism of the reduction of cells in the ventricle. In this regard, myocyte necrosis may have occurred in a scattered fashion, involving single or small groups of cells. Such a phenomenon would lead to local expansions of the interstitial space, which would be difficult to detect through relative volume measurements. This type of adaptation has been observed previously with aging in rats³⁸ and humans.²² Importantly, rapid ventricular tachycardia may have resulted in the activation of programmed cell death in the myocardium, and this mechanism may have contributed to the overall loss of myocytes in the ventricle. Programmed myocyte cell death has been documented recently during postnatal development of the heart in vivo⁴⁵ and after ischemic reperfusion injury in vitro.⁴⁶ Apoptotic cell death typically involves individual cells, and, in contrast to necrotic myocyte cell death, inflammatory reaction, fibroblast stimulation, vascular proliferation, and scar formation are absent.^{47 48 49} This possibility may shed light on some of the apparent contrasting results obtained with pacing-induced dilated cardiomyopathy.^{11 14 16 18} Ventricular dilation cannot occur in the absence of both cell death and cellular hypertrophy.⁵⁰

Myocyte Cellular Hypertrophy and Ventricular Wall Stress
Findings in this investigation indicate that myocyte cellular hypertrophy occurred as a result of chronic severe tachycardia, and cellular enlargement was capable of compensating for myocyte cell loss. However, this cellular reactive response did not normalize diastolic wall stress, which remained markedly elevated. Since the aggregate amount of myocytes in the ventricle decreased 5% and intracavitary volume increased 62%, the ventricular muscle mass–to–chamber volume ratio decreased 41% with rapid pacing. These anatomic modifications are consistent with ventricular shape changes characteristic of decompensated eccentric hypertrophy,⁸ the form of hypertrophy that distinguishes the transition from compensated pressure, and/or volume-overload hypertrophy to myocardial dysfunction and end-stage cardiac failure.^{51 52} An increasing pressure load induces concentric ventricular hypertrophy in which wall thickness increases without chamber enlargement, leading to an augmentation in myocardial mass-to-volume ratio. On the other hand, an increased volume load typically induces enlargement of the ventricular chamber volume and a corresponding expansion in myocardial tissue volume to retain a constant muscle mass–to–chamber volume ratio.^{8 51 52} These factors characterize eccentric hypertrophy in the well-balanced stage in which chamber dilatation is accompanied by a proportional increase in wall thickness, so the ratio of wall thickness to chamber radius is not altered.⁸ When these relations are not preserved, decompensated eccentric hypertrophy develops, as observed here in the cardiomyopathic heart. In accordance with previous results,^{8 19 21 24 38} diastolic Laplace overloading may represent the prevailing mechanical stimulus implicated in the growth response of the myocardium in this setting.

The dimensional changes of ventricular myocytes in pacing-induced dilated cardiomyopathy involved a prevailing increase in cell length rather than in diameter, resulting in a significant reduction in the cell diameter–to–cell length ratio. Such a modification in myocyte shape is the direct consequence of the magnitude of increase in diastolic load at the cellular level. Moreover, this configurational alteration constituted one of the major mechanisms of ventricular dilation after pacing. These observations are consistent with previous findings in the pig model.^{11 17} The possibility also may be raised that architectural rearrangement of myocytes with side-to-side slippage of muscle cells may have contributed to the expansion in cavitary volume.⁵³ Side-to-side slippage of myocytes would generate a reduction in the mural number of cells, wall thinning, and chamber enlargement. This pattern of ventricular dilation was not investigated here, but it has been inferred previously in wall thickness changes accompanying variations of ventricular volume in the intact heart in vitro⁵⁴ after ventricular dysfunction,⁵⁵ myocardial ischemia,⁵⁶ infarction,⁵³ and during early postnatal development.⁵⁷

Myocyte Cell Volume and Number
In the last three decades, several methodologies have been described for the evaluation of myocyte size and number in the mammalian heart. These techniques involved the quantitative analysis of tissue sections,^{7 19 20 21 22 23 53 57} the evaluation of enzymatically dissociated myocytes,^{11 15} or a combination of both.^{38 58} The morphometric procedure offers an in situ approach that yields reliable information on the volume composition of the myocardium and number of myocyte nuclei in the tissue.^{19 20 21 22 23} In contrast, myocyte isolation provides accurate data on the frequency distribution of nuclei per cell.^{11 38 58} By use of both techniques, the numbers of mononucleated, binucleated, trinucleated, and tetranucleated myocytes in the ventricle have been calculated.^{38 58} However, the estimation of the volume fraction of myocardium occupied by each myocyte population has been problematic. Similarly, the determination of the average cell volume of mononucleated and multinucleated myocytes has never been obtained. The lack of these parameters has made difficult the characterization of the hypertrophic reaction of the different myocyte populations of the heart under any condition of overload. These limitations have been overcome by the technique described in this study which combined an in situ morphometric analysis with the determination of cell configuration and dimensions of isolated myocytes. By the use of confocal microscopy, three-dimensional optical reconstruction of single myocytes could be obtained. On this basis, the contribution of each myocyte population to the overall muscle mass of the ventricular myocardium was determined. Moreover, the volumes of mononucleated and multinucleated cells and their changes with rapid ventricular pacing were established.

Limitations of the Study and Conclusions
There are several limitations in the current investigation that must be acknowledged. (1) The number of dogs was relatively small, and age differences among the animals could not be carefully evaluated. (2) The measurements of myocyte size, shape, and number could not be obtained in the same hearts. (3) The necessity of isolating ventricular myocytes by enzymatic digestion may have affected the accuracy of sampling. Smaller mononucleated myocytes or larger multinucleated cells may have been preferentially preserved during isolation, so that samples collected may not be representative of the actual proportion of cell populations in the tissue. (4) The distribution of diastolic wall stress from the apex to the base of the heart was not directly measured but was computed by combining hemodynamic estimations obtained in vivo with anatomic parameters determined in the diastolic arrested perfusion fixed heart. This is particularly important because the effects of these manipulations necessary for the fixation and histological analysis of the myocardium may have altered significantly the in vivo properties of the heart. Although these limitations have to be considered, the results of the present study suggest that the cardiomyopathic heart associated with rapid ventricular pacing is characterized by myocyte cell loss, reactive myocyte cellular hypertrophy, and reduction in ventricular mass–to–chamber volume ratio. Whether myocyte loss is the prevailing etiological factor responsible for the development of decompensated eccentric hypertrophy in this model requires further investigation.

	Acknowledgments

 
This work was supported by grants HL-38132, HL-39902, HL-4056, HL-50142, and PO-1-HL-43023 from the National Heart, Lung, and Blood Institute. The expert technical assistance of Maria Feliciano is greatly appreciated. Dr Ervin Szoke was a visiting fellow from the Semmelweis University, Budapest, Hungary.

Received March 3, 1995; revision received April 26, 1995; accepted May 18, 1995.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Dec GW, Fuster V. Idiopathic dilated cardiomyopathy. N Engl J Med. 1994;331:1564-1574. [Free Full Text]
   
2. Johnson RA, Palacios I. Dilated cardiomyopathies of the adult. N Engl J Med. 1982;307:1051-1058. [Medline] [Order article via Infotrieve]
   
3. Schwarz F, Mall G, Zebe H, Blickle J, Derks H, Manthey J, Kubler W. Quantitative morphologic findings of the myocardium in idiopathic dilated cardiomyopathy. Am J Cardiol. 1983;51:501-506. [Medline] [Order article via Infotrieve]
   
4. Unverferth DV, Baker PB, Swift SE, Chaffee R, Fetters JK, Uretsky BF, Thompson ME, Leier CV. Extent of myocardial fibrosis and cellular hypertrophy in dilated cardiomyopathy. Am J Cardiol. 1986;57:816-820. [Medline] [Order article via Infotrieve]
   
5. Roberts WC, Siegel RJ, McManus BM. Idiopathic dilated cardiomyopathy: analysis of 152 necropsy patients. Am J Cardiol. 1987;60:1340-1355. [Medline] [Order article via Infotrieve]
   
6. Tanaka M, Fujiwara H, Onodera T, Wu DJ, Matsuda M, Hamashima Y, Kawai C. Quantitative analysis of narrowings of intramyocardial small arteries in normal hearts, hypertensive hearts, and hearts with hypertrophic cardiomyopathy. Circulation. 1987;75:1130-1139. [Abstract/Free Full Text]
   
7. Beltrami CA, Finato N, Rocco M, Feruglio GA, Puricelli C, Cigola E, Sonnenblick EH, Olivetti G, Anversa P. The cellular basis of dilated cardiomyopathy in humans. J Mol Cell Cardiol. 1995;27:291-305. [Medline] [Order article via Infotrieve]
   
8. Anversa P, Olivetti G, Meggs LG, Sonnenblick EH, Capasso JM. Cardiac anatomy and ventricular loading after myocardial infarction. Circulation. 1993;87(suppl VII):VII-22-VII-27.
   
9. Armstrong PW, Stopps TP, Ford SE, DeBold AJ. Rapid ventricular pacing in the dog: pathophysiologic studies of heart failure. Circulation. 1986;74:1075-1084. [Abstract/Free Full Text]
   
10. Shannon RP, Komamura K, Stambler BS, Bigaud M, Manders T, Vatner SF. Alterations in myocardial contractility in conscious dogs with dilated cardiomyopathy. Am J Physiol. 1991;260:H1903-H1911. [Abstract/Free Full Text]
    
11. Spinale FG, Zellner JL, Tomita M, Crawford FA, Ziles MR. Relation between ventricular and myocyte remodeling with the development and regression of supraventricular tachycardia-induced cardiomyopathy. Circ Res. 1991;69:1058-1067. [Abstract/Free Full Text]
    
12. Kiuchi K, Shannon RP, Komamura K, Cohen DJ, Bianchi C, Homcy CJ, Vatner ST, Vatner DE. Myocardial ß-adrenergic receptor function during the development of pacing-induced heart failure. J Clin Invest. 1993;91:907-914.
    
13. Shannon RP, Komamura K, Shen YT, Bishop SP, Vatner SF. Impaired regional subendocardial coronary flow reserve in conscious dogs with pacing-induced heart failure. Am J Physiol. 1993;265:H801-H809. [Abstract/Free Full Text]
    
14. Komamura K, Shannon RP, Ihara T, Shen YT, Mirsky I, Bishop SP, Vatner SF. Exhaustion of Frank-Starling mechanism in conscious dogs with heart failure. Am J Physiol. 1993;265:H1119-H1131. [Abstract/Free Full Text]
    
15. Zellner JL, Spinale FG, Eble DM, Hewett KW, Crawford FA. Alterations in myocyte shape and basement membrane attachment with tachycardia-induced heart failure. Circ Res. 1991;69:590-600. [Abstract/Free Full Text]
    
16. Komamura K, Shannon RP, Pasipoularides A, Ihara T, Lader AS, Patrick TA, Bishop SP, Vatner SF. Alterations in left ventricular diastolic function in conscious dogs with pacing-induced heart failure. J Clin Invest. 1992;89:1825-1838.
    
17. Spinale FG, Crawford FA, Hewett KW, Carabello BA. Ventricular failure and cellular remodeling with chronic ventricular tachycardia. J Thorac Cardiovasc Surg. 1991;102:874-882. [Abstract]
    
18. Weber KT, Pick R, Silver MA, Moe GW, Janicki JS, Zucker IH, Armstrong PW. Fibrillar collagen and remodeling of dilated canine left ventricle. Circulation. 1990;82:1387-1401. [Abstract/Free Full Text]
    
19. Anversa P, Zhang X, Li P, Capasso JM. Chronic coronary artery constriction leads to moderate myocyte loss and left ventricular dysfunction and failure in rats. J Clin Invest. 1992;89:618-629.
    
20. Beltrami CA, Finato N, Rocco M, Feruglio GA, Puricelli C, Cigola E, Quaini F, Sonnenblick EH, Olivetti G, Anversa P. Structural basis of end-stage failure in ischemic cardiomyopathy in humans. Circulation. 1994;89:151-163. [Abstract/Free Full Text]
    
21. Anversa P, Palackal T, Sonnenblick EH, Olivetti G, Capasso JM. Hypertensive cardiomyopathy: myocyte nuclei hyperplasia in the mammalian heart. J Clin Invest. 1990;85:994-997.
    
22. Olivetti G, Melissari M, Capasso JM, Anversa P. Cardiomyopathy of the aging human heart: myocyte loss and reactive cellular hypertrophy. Circ Res. 1991;68:1560-1568. [Abstract/Free Full Text]
    
23. Olivetti G, Melissari M, Balbi T, Quaini F, Sonnenblick EH, Anversa P. Myocyte nuclear and possible cellular hyperplasia contribute to ventricular remodeling in the hypertrophy senescent heart in humans. J Am Coll Cardiol. 1994;24:140-149. [Abstract]
    
24. Anversa P, Li P, Malhotra A, Zhang X, Herman M, Capasso JM. Effects of hypertension and coronary constriction on cardiac function, morphology, and contractile proteins in rats. Am J Physiol. 1993;265:H713-H724. [Abstract/Free Full Text]
    
25. Wang J, Seyedi N, Xu XB, Wolin MS, Hintze TH. Defective endothelium-mediated control of coronary circulation in conscious dogs after heart failure. Am J Physiol. 1994;266:H670-H680. [Abstract/Free Full Text]
    
26. Mathew R, Wang J, Gewitz MH, Hintze TH, Wolin MS. Congestive heart failure alters receptor-dependent cAMP-mediated relaxation of canine pulmonary arteries. Circulation. 1993;87:1722-1728. [Abstract/Free Full Text]
    
27. Hintze TH, Wang J, Seyedi N, Wolin MS. Role of EDRF/NO in chronic coronary blood flow states during myocardial dysfunction and failure. In: Bevan JA, Kaley G, Rubanyi GM, eds. Flow-Dependent Regulation in Cardiovascular Function. New York/London: Oxford