Effects of Captopril on Contractility After Myocardial Infarction: Experimental Observations Sheldon E. Litwin,

MD,

Thomas E. Raya, MD, Alberta Warner, and Steven Goldman, MD

After large myocardial infarction, compromised left ventricular (LV) function and changes in the peripheral circulation result in the syndrome of chronic congestiie heart failure. Although treatment with angitiensin-converting enzyme inhibitors improve cardiovascular function, it is diicult to determine whether thii benefit is due to changes in organ versus muscle functiin. The rat model of heart failure, created by ligating the left coronary artery, results in pathophysiology that is similar to that seen in patkmts, i.e., increased LV end-diastolic: pressure and volume, hypettrophy of the noninfarcted myocardium, prolongation of the time constant of LV relaxation, decreased venous compliance, and increased total blood volume. In noninfarcted papillary muscles, isolated from rats with heart failure, maximal developed tension and peak rate of tension rise (sdT/c/t) are decreased, time to peak tension is prolonged, and myocardial stiiess is increased. Morphologic changes include an increase in papillary muscle myocyte cross-sectional area and an increase in myocardial hydroxyproline content. Captopril(2 g/liter drinking water) atters LV loading by decreasie arterial pressure, increasing venous compliance, and decreasing blood volume. This results in a decrease in LV end-diastolii pressure and volume. In the noninfarcted myocardium, time to peak tension is shortened,

MD,

Christine M. Litwin,

MD,

whereas developed tension, +dT/dt, and muscle sttt remain abnormal. Captopril decreases myocyte cross-sectional area, but collagen content remains elevated. Thus, in the rat infarct model of heart failure, treatment with captopril alters LV remodeling and hypertrophy but produces only modest improvement in muscle function. In thii model, after treatment for 3 weeks, it appears that the major benefit from angiotensinconverting enzyme inhibition is due to improvements in organ function resulting from changes in loadiN condiiins, whereas muscle function remains compromised. (Am J Cardiil1991;68:26D44D)

L

arge myocardial infarction frequently produces progressive left ventricular (LV) dilation and hypertrophy of myocytes in the noninfarcted myocardium. These changes in LV geometry, referred to as LV remodeling, may contribute to development of the syndrome of chronic congestive heart failure. Laboratory and clinical studies have shown that treatment of heart failure with angiotensin-converting enzyme (ACE) inhibitors, such as captopril, can improve LV function and survival.‘-9 Although it is commonly believed that the beneficial effects of captopril are due to improvement in ventricular loading conditions, this explanation may be too simplistic. We hypothesized that following large myocardial infarcFrom the Departments of Internal Medicine and Pathology, tion, functional impairment of noninfarcted myoTucson Veterans Administration Medical Center, and the Univercardium would occur due to the increased load sity of Arizona Heart Center, Tucson, Arizona. Supported by grants from the Arizona Afhliate of the American Heart Associa- borne by the residual myocytes. Unloading these tion, National American Heart Association, Arizona Disease myocytes might be expected to result in a gradual Control Research Commission, Veterans Administration, and the improvement in their function. Because of the National Institutes of Health (HL-20984). The data in this article have been published previously in part in Circulation (1988;77:1424- complex effects of simultaneous changes in myocardial function, ventricular geometry, and ventricular1431) Circulation Research (1989;64:330-338) and Circulation (1991;83:1028-1037). vascular coupling on global LV function, it is Address for reprints: Steven Goldman, MD, Cardiology lllC, difficult to assessthe relative importance of changes Tucson Veterans -4dministration Medical Center, Tucson, Arizona in each area. This report focuses on our efforts to 85723.

26D

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 68

NOVEMBER 18, 1991

diographic (ECG) recording. Previous studies have shown that most rats identified in this manner have infarction of greater than 40% of the left ventricle, elevated LV end-diastolic pressure and volume, and impaired systolic function.8,93’3-‘5 Rats with large infarctions and sham-operated rats were randomly assigned to no treatment or treatment with captopril (2 g/liter; ER Squibb, Princeton, NJ). Treatment with captopril was started 21 days after surgery and was continued for 21 days. This time course was chosen because histologic healing of the infarct region is complete at 21 days. To obtain pressure measurements, the rats were anesthetized with methoxyflurane, and a micromanometer-tipped catheter (Millar Instruments, Houston, TX) was inserted into the right carotid artery. The catheter was advanced into the aorta and then into the left ventricle for recording of pressures. With the rat lightly anesthetized and breathing spontaneously, recordings were made on a physiologic recorder (model 2400, Gould Instrument Co., Cleveland, OH). This catheter was then exchanged for a polyethylene catheter. Using techniques previously described, mean circulatory filling pressure, venous compliance, and blood volume were measured.** In brief, transient circulatory arrest was produced by inflation of a balloon in the right atrium. Mean circulatory filling pressure was calculated from the venous and arterial plateau pressures. This procedure was repeated after 210% changes in blood volume. Venous compliance was calculated from the slope of the total vascular pressure-volume relation. Pressure-volume data from the isolated left ventricle were recorded using methods previously described.“~” These measurements were made in the excised left ventricle with a timed infusion of normal saline. Isometric muscle tunctlon: Rats were killed by decapitation, the heart rapidly excised, and the noninfarcted posterior papillary muscle was dissected free.14The muscle was suspended vertically from an isometric force transducer (Metrigram Mtm, Gould Instrument Co.) and isometric funcMETHODS All studies described in this article conformed tion and passive myocardial stiffness were meato the animal care guidelines of the American sured.14 Hlstologk stud& After the papillary muscle Physiological Society. Myocardial infarction was produced in male Sprague-Dawley rats (Harlan, was removed, the heart was separated into right Indianapolis, IN) weighing 22&260 g, by ligation ventricle and left ventricle plus septum, weighed, of the proximal left coronary artery.“-13 Rats were and immersion-fixed in 10% buffered formalin. screened for evidence of large myocardial infarc- Thin sections of the ventricle were stained with tion 21 days after surgery using surface electrocar- Masson’s trichrome and infarct size was measured

separate alterations in muscle function from those in organ function and how captopril affects each of these in rats with heart failure. It is difficult to evaluate intrinsic myocardial function following myocardial infarction by measuring global ventricular function because these indexes of function represent the net change in the residual noninfarcted myocardium and in the scar. For example, it is important to define changes in wall stress to understand alterations in both systolic and diastolic function. Traditional methods of measuring myocardial wall stress in vivo rely on mathematical models that assume that the ventricle has symmetrical shape, uniform wall thickness, homogeneous myocardial elasticity, and a constancy of forces from epicardium to endocardium.” These assumptions cannot be applied to the infarcted heart. Therefore, to investigate this problem in the research laboratory, we chose to perform studies using isolated myocardial tissue as well as intact rats to evaluate adequately the effects of ACE inhibition on ventricular loading conditions and myocardial mechanics. Our unifying hypothesis in these studies was that activation of systemicor tissue renin-angiotensin systemsmay play a key role in the sequence of events leading to heart failure. Because of direct myocardial effects of angiotensin II, such as trophism and inotropism, in addition to its effects on the peripheral circulation, we reasoned that blocking the production of angiotensin II might interrupt or slow the process of LV remodeling and the development of heart failure after myocardial infarction. We hypothesized that the effectivenessof captopril in improving organ function is dependent on its ability to cause both arterial and venous dilation. We also hypothesized that myocardial infarction would result in depressed contractility and increased muscle stiffness in the noninfarcted myocardium. We speculated that treatment with captopril would partially reverse these abnormalities, since these parameters were, at least in part, load dependent.

A SYMPOSIUM: VENTRICULAR REMODELING

27D

by tracing the outline of the infarcted and noninfarcted regions of the ventricle at 4 different levels. Infarct size is reported as the mean percent of epicardial and endocardial circumference occupied by scar tissue for the 4 sections. Papillary muscles were processed as noted above and cross-sectional slices were made at 3 different levels. Cross-sectional areas of approximately 100 myocytes were measured for each papillary muscle.14The operator was blinded to the experimental group during the analysis. Myocardial hydroxyproline content was measured using previously described methods.14Briefly, the scar was removed from the left ventricle and the tissue was frozen. At a later time, the tissue was dried, weighed, and hydrolyzed in 6 N HCl at 105°C. Hydroxyproline content of the samples was determined using spectrophotometry at 557 nm. Statistical analysis: Values are reported as mean k standard error of the mean. Muscle stiffness constants, regression coefficients, and values for venous compliance were determined by linear regression using the method of least squares. Isometric parameters are means of 3-5 twitches and muscle stiffness values are the mean of at least 3 stretches for each muscle. Differences between groups were detected with l-way analysis of variance and intergroup comparisons were performed using the Student-Newman-Keuls test or Dunnett’s test for multiple comparisons against a single control.

A.

--s 2 200 -loo--

f

400

300

0. B.

400

;;; 300 5 iz 200 ~ 100

t

0.

C. 1oooT

‘;;; .s

RESULTS Hemodynamic studies: The baseline chamber weights, heart weights and hemodynamics are shown in Figure 1. The rats with heart failure after myocardial infarction had increased right ventricular (RV) weights, increased RV/body weight, but no change in LV weight/body weight. In the captopril-treated rats body weight was unchanged while LV weight, RV weight, and RV/body weight were all decreased compared with untreated rats with large infarctions. The hemodynamic changes are shown in Figure 2. In the rats with heart failure there was no change in heart rate, a decrease in mean aortic pressure and an increase in LV end-diastolic and right atria1 pressures. Hematocrit was unchanged, mean circu-

600

‘“:i: D. 3.02.5 -2.0 -: 1

1.5.-

: 1.0~0.5 --

4

0.0-l CONTROL

2BD

INFARCT

INFARCT + CAPTO

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 68

FIGURE 1. Changes in bedy weight (BW) (A), right ventrkular (RV) weight (B), left ventricular (LV) weight (C) and left ventrkular weight/body weight (DJ. Values are meen -+ standard error of the mean. *p < 0.05, control versus infarct (n = 11); #p < 0.05 infarct versus infarct-ceptoprtl (CAPTO) (n = 9). Note that comparisons were not made between control versus infarct-captopril.

NOVEMBER 18, 1991

latory filling pressure and blood volume were increased, and venous compliance was decreased in the rats with heart failure compared to control (Figure 3). Treatment of rats with heart failure with captopril did not change heart rate, further decreased mean aortic pressure and LV enddiastolic pressure (Figure 2). In the captopriltreated rats with large infarctions, hematocrit was unchanged, mean circulatory filling pressure and blood volume decreased, and venous compliance increased. The end result of these hemodynamic changes was that in the rats with heart failure, the passive LV pressure-volume relation was shifted away from the pressure axis, indicating LV dilation. Captopril treatment in the infarct rats did not shift the pressure-volume curve but did decrease operating LV end-diastolic pressure and vo1ume.8 Papillary muscle studies: In 11 rats with large infarctions and heart failure, there were no changes in papillary muscleweight (7.4 + 0.6 mgvs 6.9 f: 0.7 mg) or muscle cross-sectional area (1.2 + 0.1 mm2 vs 1.0 2 0.1 mm’), but there was a decrease (p < 0.05) in the length associated with maximum developed tension (5.6 + 0.4 mm vs 7.1 + 0.4 mm). Myocyte cross-sectional area was increased (Figure 4) and collagen content, defined by changes in hydroxproline levels in the noninfarcted myocardium, was increased (8.5 + 0.9 pg/mg dry weight vs 3.3 +- 0.3 pg/mg dry weight; p