Progressive ventricular remodeling in rat with myocardial infarction JANICE M. PFEFFER, EUGENE BRAUNWALD
MARC A. PFEFFER, (With the Technical
PETER J. FLETCHER, AND Assistance of Cynthia R. Steinberg
and Peter Finn)
Cardiovascular Division, Department of Medicine, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts 02115
PFEFFER, JANICE, M., MARC A. PFEFFER, PETER J. FLETCHER, AND EUGENE BRAUNWALD. Progressive ventricular Am. J. Physiol. remodeling in rat with myocardial infarction.
260 (Heart Circ. Physiol. 29): H1406-H1414, 1991.-Ventricular dilatation may have important prognostic implications for the survival of patients with left ventricular (LV) dysfunction. To determine the manner and extent to which the left ventricle of the rat remodels and dilates after myocardial infarction, we obtained the passive pressure-volume relationships, chamber stiffness constants, and mass during both the early and late phases. In moderate and large infarcts as inflammation and edema developed, LV weight increased then progressively decreased as a thin scar formed, returning to normal values as a result of compensatory hypertrophy of the residual myocardium. LV dilatation occurred in all rats with infarcts but to different extents depending on infarct size and duration. In the early postinfarction phase, pressure-volume relationship was relatively unchanged in all infarct-size groups, except for significant rightward shift in low pressure range for rats with moderate and large infarcts and significant leftward shift in high pressure range for rats with small infarcts. During resolution of the inflammatory response, LV dilatation occurred in all infarct groups in relation to infarct size. As scar formation became complete, LV enlargement did not progress in rats with small infarcts but did so in rats with moderate and large infarcts. LV chamber stiffness remained within the range of normal values during the early phase in all rats with infarcts but decreased significantly during the late phase in rats with moderate and large infarcts in association with the extent of ventricular enlargement. Alterations in the volume-to-mass ratio (V/V,) were most marked in the late postinfarction phase, wherein both volume (increased) and mass (decreased, then increased) changed dramatically and V/V, progressively increased in rats with large infarcts. pressure-volume relationship; ventricular dilatation; ventricular hypertrophy; congestive heart failure THE HEART ADAPTS to the sustained imposition of a functional overload by an increase in its mass and/or cavitary volume, the pattern of which depends on the type of load imposed. The loss of contractile tissue consequent to a myocardial infarction places an augmented load on the residual myocardium, which may use its systolic (increase in contractility) and/or diastolic (Frank-Starling mechanism) reserves to maintain systemic perfusion. The chronic use of these compensatory mechanisms may result in concurrent or disparate changes in ventricular mass, volume, and compliance, H1406
0363-6135/91
$1.50
which depend on the extent of the initial damage to the myocardium, and thereby ventricular dysfunction, and which may eventually become mechanically disadvantageous, deleteriously affecting ventricular performance. Previous studies of experimental myocardial infarction in dogs have reported no change (8, 29) or a decrease (12,X3) in left ventricular end-diastolic volumes. In most of these studies infarct size was relatively small (ll-28% of the left ventricle) and ventricular performance was minimally altered (12). Coronary artery ligation in the rat produces a broad spectrum of ventricular dysfunction, ranging from minimal impairment to overt heart failure, depending on the size of myocardial infarction (20). In a previous study of rats with healed (26 days) myocardial infarction, we found an infarct-size-related increase in left ventricular volume that occurred primarily at low (~2.5 mmHg) filling pressures, as the pressure-volume relationships >2.5 mmHg were shifted in parallel to the right (7). To determine the manner and extent to which the left ventricle is remodeled throughout the process of healing, or scar formation, and beyond, we obtained the passive pressure-volume relationships of the infarcted left ventricle of the rat at various times during both the early and late phases of ventricular infarction. Because infarct size is an important determinant of remodeling, rats with infarcts of varying sizes were studied. METHODS
Myocardial infarctions were produced in 4- to 6-moold female normotensive Wistar rats (West Jersey Biological SUPPlY) bY a previous1 .y described method of coronary artery ligati .on (7, 20). All rats then were housed under identical conditions in a 12-h light-dark cycle and given food and water ad libitum. Pressure-volume relationship. To assess the potential changes in ventricular cavitary size, mass, and passive mechanical properties after i.nfarction we studied rats at six preselected time periods after coronary artery ligation: 4-6 h and 1, 2, 7, 19, and 106 days. These time intervals were chosen to encompass both the early phase of acute myocardial necrosis and inflammation (4-6 h, 1 and 2 days) and the later phase of fibrous tissue replacement and histological healing (7, 19, and 106 days). After hemodynamic studies to be reported separately, the heart was arrested in diastole with potassium chloride and pressure-volume curves were generated by the infu-
Copyright 0 1991 the American Physiological
Society
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on August 14, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
VENTRICULAR
REMODELING
AFTER
sion of saline over a pressure range of O-30 mmHg, a method reported in detail previously (7). Ventricular volumes were determined at every unit millimeter of mercury of pressure on the pressure-volume curve. The pressure-volume data from 2.5 to 30 mmHg were fitted by a single two-constant exponential (P = bekv) and the value of k was used as an index of overall left ventricular chamber stiffness (7). After inscription of the pressure-volume curves, a volume of ice-cold saline (hearts studied at 4-6 h after coronary artery ligation) or Formalin (hearts studied at 24 h or longer) corresponding to that required to establish a pressure of 5 mmHg was infused into the left ventricle. The right and left ventricles then were separated and weighed immediately for hearts studied at 4-6 h and after 24 h of fixation for hearts studied 24 h or longer after ligation. Infarct size measurement. Three different histological techniques were necessary to detect the extent of necrotic myocardium or scar tissue depending on the time after coronary artery ligation. Hearts studied between 4 and 6 h after coronary artery ligation were processed immediately and stained with triphenyltetrazolium chloride (6). Polaroid photographs of each section (magnification X10) were taken, and the outlines of stained and unstai ned tissue observed from each section were drawn on the photographs for subsequent planimetry. For hea rts studied between 1 and 7 days after coronary artery ligation, only minor fibrotic changes were present and staining with hematoxylin alone permitted the visualization of nuclear changes and the delineation of the area of necrosis fo r planimetry. In hearts studied ~2 wk after infarction, fibrous infarct was detected visually bY a modified Masson stain as we ha ,ve descri .bed prev iously
(7)
The lengths of scar (or of necrotic tissue) and of noninfarcted muscle for both the endocardial and epicardial surfaces of each histological section were determined by planimetry of the Polaroid photographs (4- to 6-h group) or of the projected histological slides (all other groups). The lengths of scar for the endocardial and epicardial surfaces for all histological sections were numerically summed separately as were the endocardial and epicardial circ umferences. T he ratio of the sums of the lengths of scar and of surface circumfere rices defined the infarct size for each of the myocardial surfaces. Final infarct size was expressed in a percentage as the average of the infarct sizes of the endocardial and epicardial surfaces times one hundred. Statistical analysis. Results were expressed as means 2 SE. For each dependent variable, the initial analysis was performed as described by Gill (9) for multiple independent groups with two treatment effects, a possible interaction, and an unbalanced replication. Total variance was partitioned into that associated with the two main effects of infarct-size groupings [3 degrees of freedom (df)] and time intervals (5 df), their possible interaction (15 df), and pooled within group error mean square. When there was a significant interaction, the analysis was repeated separately for the “early” time intervals (4-6 h, 1 and 2 days) and for the “late” time intervals (7, 19, and 106 days). In general, there were no
MYOCARDIAL
H1407
INFARCTION
significant interactions at the early time intervals, indicating that the infarcted groups behaved uniformly during the acute phase of myocardial infarction; however, there were highly significant interactions for the late time intervals for most variables, indicating that the infarcted groups behaved differently with time in the more chronic phase of myocardial infarction. To detect differences between the infarcted groups and the corresponding noninfarcted control groups, a one-way analysis of variance (with Dunnett’s test for 3 comparisons) then was performed at each time interval; significant differences between a given infarct-size group and the noninfarcted control group at each time interval are presented in the tables and figures. To detect differences that occurred over time within each infarct-size group, a one-way analysis of variance (with Duncan’s multiplerange test for significant differences) was performed for each group; significant differences between time intervals for a given infarct-size group are presented in the text. RESULTS
A total of 279 rats completed the protocol for passive pressure-volume relationships (Table 1). For the purpose of data analysis and comparison, rats were classified as those with small infarcts (~5 and ~30% of the left ventricular circumference), moderate infarcts (230 and
MARC A. PFEFFER, (With the Technical
PETER J. FLETCHER, AND Assistance of Cynthia R. Steinberg
and Peter Finn)
Cardiovascular Division, Department of Medicine, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts 02115
PFEFFER, JANICE, M., MARC A. PFEFFER, PETER J. FLETCHER, AND EUGENE BRAUNWALD. Progressive ventricular Am. J. Physiol. remodeling in rat with myocardial infarction.
260 (Heart Circ. Physiol. 29): H1406-H1414, 1991.-Ventricular dilatation may have important prognostic implications for the survival of patients with left ventricular (LV) dysfunction. To determine the manner and extent to which the left ventricle of the rat remodels and dilates after myocardial infarction, we obtained the passive pressure-volume relationships, chamber stiffness constants, and mass during both the early and late phases. In moderate and large infarcts as inflammation and edema developed, LV weight increased then progressively decreased as a thin scar formed, returning to normal values as a result of compensatory hypertrophy of the residual myocardium. LV dilatation occurred in all rats with infarcts but to different extents depending on infarct size and duration. In the early postinfarction phase, pressure-volume relationship was relatively unchanged in all infarct-size groups, except for significant rightward shift in low pressure range for rats with moderate and large infarcts and significant leftward shift in high pressure range for rats with small infarcts. During resolution of the inflammatory response, LV dilatation occurred in all infarct groups in relation to infarct size. As scar formation became complete, LV enlargement did not progress in rats with small infarcts but did so in rats with moderate and large infarcts. LV chamber stiffness remained within the range of normal values during the early phase in all rats with infarcts but decreased significantly during the late phase in rats with moderate and large infarcts in association with the extent of ventricular enlargement. Alterations in the volume-to-mass ratio (V/V,) were most marked in the late postinfarction phase, wherein both volume (increased) and mass (decreased, then increased) changed dramatically and V/V, progressively increased in rats with large infarcts. pressure-volume relationship; ventricular dilatation; ventricular hypertrophy; congestive heart failure THE HEART ADAPTS to the sustained imposition of a functional overload by an increase in its mass and/or cavitary volume, the pattern of which depends on the type of load imposed. The loss of contractile tissue consequent to a myocardial infarction places an augmented load on the residual myocardium, which may use its systolic (increase in contractility) and/or diastolic (Frank-Starling mechanism) reserves to maintain systemic perfusion. The chronic use of these compensatory mechanisms may result in concurrent or disparate changes in ventricular mass, volume, and compliance, H1406
0363-6135/91
$1.50
which depend on the extent of the initial damage to the myocardium, and thereby ventricular dysfunction, and which may eventually become mechanically disadvantageous, deleteriously affecting ventricular performance. Previous studies of experimental myocardial infarction in dogs have reported no change (8, 29) or a decrease (12,X3) in left ventricular end-diastolic volumes. In most of these studies infarct size was relatively small (ll-28% of the left ventricle) and ventricular performance was minimally altered (12). Coronary artery ligation in the rat produces a broad spectrum of ventricular dysfunction, ranging from minimal impairment to overt heart failure, depending on the size of myocardial infarction (20). In a previous study of rats with healed (26 days) myocardial infarction, we found an infarct-size-related increase in left ventricular volume that occurred primarily at low (~2.5 mmHg) filling pressures, as the pressure-volume relationships >2.5 mmHg were shifted in parallel to the right (7). To determine the manner and extent to which the left ventricle is remodeled throughout the process of healing, or scar formation, and beyond, we obtained the passive pressure-volume relationships of the infarcted left ventricle of the rat at various times during both the early and late phases of ventricular infarction. Because infarct size is an important determinant of remodeling, rats with infarcts of varying sizes were studied. METHODS
Myocardial infarctions were produced in 4- to 6-moold female normotensive Wistar rats (West Jersey Biological SUPPlY) bY a previous1 .y described method of coronary artery ligati .on (7, 20). All rats then were housed under identical conditions in a 12-h light-dark cycle and given food and water ad libitum. Pressure-volume relationship. To assess the potential changes in ventricular cavitary size, mass, and passive mechanical properties after i.nfarction we studied rats at six preselected time periods after coronary artery ligation: 4-6 h and 1, 2, 7, 19, and 106 days. These time intervals were chosen to encompass both the early phase of acute myocardial necrosis and inflammation (4-6 h, 1 and 2 days) and the later phase of fibrous tissue replacement and histological healing (7, 19, and 106 days). After hemodynamic studies to be reported separately, the heart was arrested in diastole with potassium chloride and pressure-volume curves were generated by the infu-
Copyright 0 1991 the American Physiological
Society
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on August 14, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
VENTRICULAR
REMODELING
AFTER
sion of saline over a pressure range of O-30 mmHg, a method reported in detail previously (7). Ventricular volumes were determined at every unit millimeter of mercury of pressure on the pressure-volume curve. The pressure-volume data from 2.5 to 30 mmHg were fitted by a single two-constant exponential (P = bekv) and the value of k was used as an index of overall left ventricular chamber stiffness (7). After inscription of the pressure-volume curves, a volume of ice-cold saline (hearts studied at 4-6 h after coronary artery ligation) or Formalin (hearts studied at 24 h or longer) corresponding to that required to establish a pressure of 5 mmHg was infused into the left ventricle. The right and left ventricles then were separated and weighed immediately for hearts studied at 4-6 h and after 24 h of fixation for hearts studied 24 h or longer after ligation. Infarct size measurement. Three different histological techniques were necessary to detect the extent of necrotic myocardium or scar tissue depending on the time after coronary artery ligation. Hearts studied between 4 and 6 h after coronary artery ligation were processed immediately and stained with triphenyltetrazolium chloride (6). Polaroid photographs of each section (magnification X10) were taken, and the outlines of stained and unstai ned tissue observed from each section were drawn on the photographs for subsequent planimetry. For hea rts studied between 1 and 7 days after coronary artery ligation, only minor fibrotic changes were present and staining with hematoxylin alone permitted the visualization of nuclear changes and the delineation of the area of necrosis fo r planimetry. In hearts studied ~2 wk after infarction, fibrous infarct was detected visually bY a modified Masson stain as we ha ,ve descri .bed prev iously
(7)
The lengths of scar (or of necrotic tissue) and of noninfarcted muscle for both the endocardial and epicardial surfaces of each histological section were determined by planimetry of the Polaroid photographs (4- to 6-h group) or of the projected histological slides (all other groups). The lengths of scar for the endocardial and epicardial surfaces for all histological sections were numerically summed separately as were the endocardial and epicardial circ umferences. T he ratio of the sums of the lengths of scar and of surface circumfere rices defined the infarct size for each of the myocardial surfaces. Final infarct size was expressed in a percentage as the average of the infarct sizes of the endocardial and epicardial surfaces times one hundred. Statistical analysis. Results were expressed as means 2 SE. For each dependent variable, the initial analysis was performed as described by Gill (9) for multiple independent groups with two treatment effects, a possible interaction, and an unbalanced replication. Total variance was partitioned into that associated with the two main effects of infarct-size groupings [3 degrees of freedom (df)] and time intervals (5 df), their possible interaction (15 df), and pooled within group error mean square. When there was a significant interaction, the analysis was repeated separately for the “early” time intervals (4-6 h, 1 and 2 days) and for the “late” time intervals (7, 19, and 106 days). In general, there were no
MYOCARDIAL
H1407
INFARCTION
significant interactions at the early time intervals, indicating that the infarcted groups behaved uniformly during the acute phase of myocardial infarction; however, there were highly significant interactions for the late time intervals for most variables, indicating that the infarcted groups behaved differently with time in the more chronic phase of myocardial infarction. To detect differences between the infarcted groups and the corresponding noninfarcted control groups, a one-way analysis of variance (with Dunnett’s test for 3 comparisons) then was performed at each time interval; significant differences between a given infarct-size group and the noninfarcted control group at each time interval are presented in the tables and figures. To detect differences that occurred over time within each infarct-size group, a one-way analysis of variance (with Duncan’s multiplerange test for significant differences) was performed for each group; significant differences between time intervals for a given infarct-size group are presented in the text. RESULTS
A total of 279 rats completed the protocol for passive pressure-volume relationships (Table 1). For the purpose of data analysis and comparison, rats were classified as those with small infarcts (~5 and ~30% of the left ventricular circumference), moderate infarcts (230 and
