Beneficial effect of amrinone on the size of acute regional ischemia in isolated rabbit hearts

Beneficial

Effect of Amrinone on the Size of Acute Regional Rabbit Hearts

Ischemia

in Isolated

A.F.E. Rump, MD, R. R&en, MD, B. Sigmund, MD, J. Fuchs, MD, S. Dhein, MD, and W. Klaus, MD The effect of inotropes on myocardial ischemia is difficult to predict because they may influence the determinants of myocardial Or demand and 0s supply differently. Several PDE-inhibitors have been reported to possess antiischemic properties related to their hemodynamic and Os-sparing effects. To assess whether PDE-inhibitors also possess direct cardioprotective properties, the effects of amrinone (2.5 X 10V5 mol/L) in comparison to isoproterenol (5 x 1O-g mol/L) and ouabain (1.5 x 10e7 mol/L) were studied in isolated rabbit hearts perfused according to Langendorff at a constant pressure (70 cmHs0) and electrically driven at a constant pacing rate. Regional ischemia was induced by coronary artery ligation and quantified by epicardial NADH fluorescence. All substances significantly increased the actively developed left ventricular pressure to a similar extent

C

LINICAL STUDIES have shown that the prognosis of myocardial infarction depends on the size of the injured area.1%2Therefore, one of the most important goals of therapeutic interventions is the limitation of infarct size. Initially, the infarcted area is smaller than the ischemic zone, and in this phase of coronary occlusion, the size of the infarct expands progressively with time.3 It has been demonstrated that during this period of time various pharmacologic interventions can lead to a salvage of ischemic tissue by preventing the tissue within the occluded zone from evolving into necrotic tissue. 4,5 On the other hand, the use of inotropic agents in the setting of acute left ventricular failure precipitated by myocardial ischemia remains problematic, because it is feared that an increase in contractility may worsen the balance between the 02 demand and s~pply.~x~ However, because several factors determining the O2 demand of the myocardium (heart rate, wall tension, contractility) and the coronary blood flow (perfusion pressure, coronary resistance) are altered by inotropes, it is difficult to predict the net effect a particular inotrope will have on the balance between O2 supply and demand in a specific location of the heart.8 Findings related to the effects of inotropes on acute myocardial ischemia are, therefore, contradictory.9J” Recently, beneficial 02-sparing effects have been reported for PDE-inhibitors in the clinical setting.“-l4 These observations are supported by experimental findings showing beneficial antiischemic properties of amrinone.i5 These findings are, however, not uncontested,16 and it still remains unclear whether the beneficial effects observed in these “in vivo” examinations were solely due to changes in hemodynamics reducing the O2 demand of the myocardium. An improvement in myocardial perfusion and, thus, direct cardioprotective properties of PDE-inhibitors might also be involved in the antiischemic effects observed. Therefore, the influence of amrinone in comparison to ouabain and isoproterenol on the size of an acute regional ischemic zone in isolated rabbit hearts using quantitative NADH fluorescence was studied.

Journalof Cardiothoracic and VascularAnesthesia,

(+20%1 (P < 0.05). Coronary flow was significantly decreased by ouabain (-15%) and significantly increased by isoproterenol (+25%) and particularly by amrinone (+50%) (P < 0.05). Neither ouabain nor isoproterenol significantly changed the intensity or the distribution pattern of NADH fluorescence, whereas the size of the ischemic zone was significantly reduced by amrinone (-25%) (P c 0.05). The PDE-inhibitor amrinone was shown to possess a direct cardioprotective effect by improving myocardial perfusion and 0s supply in isolated rabbit hearts. Copyright o 1993 by W.B. Saunders Company KEY WORDS: NADH fluorescence, dioprotection, inotropy

m yocardial ischemia, car-

METHOD

Hearts were prepared and perfused according to Langendorff.” Briefly, male rabbits (White New Zealand) (1.6 to 2.0 kg body weight) were sacrificed 10 minutes after heparin administration (1,000 W/kg IV). After exsanguination and thoracotomy, the ascending aorta was cannulated towards the heart, and hearts were perfused in situ with Tyrode solution at room temperature according to Langendorff during the time of preparation, which included cannulation of the pulmonary artery and ligation of the pulmonary and caval veins. Hearts were then transferred to the perfusion system and perfused through the coronary circulation at a constant pressure of 70 cmHzO with Tyrode solution at 37°C and equihbrated with 95% 02 and 5% CO?. The Tyrode solution had the following millimolar composition: Na+ 161.11, K+ 5.36, Ca++ 1.8, Mg++ 1.05, Cl- 147.95, HCOs- 23.8, HzP04- 0.42, glucose 10.0. Left ventricular pressure was continuously measured with a pressure transducer (Trantec Model1 800, Hugo-Sachs Elektronik, March-Hugstetten, Germany) attached to a fluid-filled rubber balloon (volume of 500 uL), which was inserted via the left atrium into the left ventricle. The volume was adapted to give an end-diastolic pressure of 0 mmHg. Left ventricular pressure was measured as actively developed left ventricular pressure, ie, systolic minus end-diastolic left ventricular pressure. Global coronary flow was estimated with graded glass cylinders. Hearts were electrically paced at a constant frequency of 180 beats/min (Stimulator-S, Hugo-Sachs Elektronik, March-Hugstetten, Germany) and the pacing rate was kept constant over the experiment. For quantification of the ischemic epicardium, photographs using endogenous NADH fluorescence were taken (Fig 1) and submitted to image processing as described.t8 The excitation light was provided by a Xe-flash (100 J within 120 us, model Strobe 1001 S.O. 1, Drello, Monchengladbach, Germany), the light was filtered through an excitation filter (UGI, 0.5 mm; Schott, Mainz Germany) and directed towards the heart by a four-point quartz light guide (diameter 10 mm, Volpi, Urdorf, Switzerland). A selected pulse of the electrical stimulator was used to trigger the flash lights without delay so that the pictures were always taken during the

From the Institutfir Phatmakologie, UniversitiitKiiln, Germany. Address reprint requests to A. Rump, Institut fiir Pharmakologie, UniversitiitKiiln, Gleuelerstr. 24, D-5000 K&z, Germany. Copyright 0 I993 by W.B. Saunders Company 1053-077019310705-0015$03.0010

Vol7. No 5 (October), 1993: pp 573-578

573

RUMP ET AL

574

PERFUSION

-

SYSTEM

CAMERA

LIGATURE LIGHT GUIDE

EX

FLASH 100

same phase of the cardiac contraction cycle (late ventricular diastole). NADH fluorescence was recorded in a darkened room on high sensitivity film (Polaroid type 667, 36 DIN) with a cut-on emission filter (OG 435/3 mm, Schott) in front of a Rolleiflex SL66 camera (shutter open) fitted with an 80 mm retro lens and bellow attachment, which allowed an image magnification of up to 1.5. After digitizing the pictures by video A/D conversion (C-1,000, Hamamatsu, TV Co, Hamamatsu, Japan) into a matrix of 256 x 256 pixels and 256 grey levels, image processing was performed on an Apple IIe microcomputer. After coronary ligation, enhanced NADH fluorescence was shown by an increase in the number of pixels with higher grey values. The number of pixels with enhanced NADH fluorescence after coronary ligation was proportional to the size of the ischemic area. Hearts were allowed to stabilize for 45 minutes after preparation before a branch of the left coronary artery was occluded for a period of 120 minutes (Fig 2). Then, the ischemic zone was reperfused for 20 minutes. Fluorescence photographs were taken as indicated in Fig 2. The area of the ischemic zone after an ischemic period of 30 minutes was taken as a reference (100%). The application of inotropes in the perfusion buffer was started immediately after the reference photography had been taken and was continued over the course of the experiment. Amrinone (Sigma, St. Louis, MO) was applicated at a concentration of 2.5 x 10m5 mol/L (4.7 p,g/mL), lying in the therapeutic serum level range. l9 For comparison, equieffective concentrations of ouabain (Serva, Heidelberg, Germany) (1.5 x lo-’ mol/L) or isoprenaline (Sigma) (5 x 1O-9 mol/L) were used. Results are given as mean 2 SEM of n experiments. Significance of the results at 95% confidence intervals was evaluated by one-way or two-way ANOVA and Scheffe test using actual numerical values. RESULTS

In all groups coronary ligation significantly (P < 0.05) decreased global coronary flow (CF) (Fig 3, Table 1) and

- FILTER Fig 1. Experimental setup for UV-flash photography. EXC-filter, excitetfon fifter (UG 1, Schott); EMfilter, emission filter (GG 435, Schott). (e Intemational Anesthesia Res Society 1993.)

J

left ventricular pressure (LVP) (Fig 4, Table l), which both increased upon reperfusion. Isoproterenol induced a significant but only temporary increase of LVP (Fig 4) with a maximum at t = 40 minutes (P < 0.05). CF was also significantly increased with a peak value at t = 40 minutes to slightly decrease thereafter, but it remained significantly elevated compared to its value before isoproterenol application (P < 0.05) (Fig 3). Ouabain induced a slower but significant and sustained increase of LVP (P < 0.05) (Fig 4). The maximal increase of LVP by ouabain was not significantly different from the peak elevation induced by isoproterenol (P > 0.05). CF was rapidly and significantly decreased by ouabain, and slowly but significantly increased again after a minimum at t = 40 minutes (P < 0.05) (Fig 3). Amrinone induced a significant elevation in LVP (P < 0.05) (Fig 4) that was slightly but not significantly superior to that observed after the application of ouabain or the maximum increase by isoproterenol (P > 0.05). The

Amrinone

I

I

I 0

t

1

I

lschemia

Equilibration

Rep.

, 30

I 120

ttt

t

NASH-Fluorescence Fig 2.

Photographs

Experimental protocol.

tt

min

EFFECTS OF AMRINONE ON MYOCARDIAL ISCHEMIA

575

Global CoronaryFlow lrchemia Ouabain ltoplsnaline Amrinone Control A%

+OO-

Fig 3. Time course of the relativs changes of the global coronary flow over the experiment as shown in the experimental protocol in Fig 2. Coronaw ligation was set at the time t = d rnktes and the respective values at t = 30 minutes were used as a reference (100%). Points represent the mean + SEM, N = 6 in each group.

O-

-10s

llma lmln)

increase in CF by amrinone was, however, significantly higher than after isoproterenol application (P < 0.05) (Fig 3). Whereas CF slightly but significantly decreased again after t = 90 minutes (P < 0.05), no significant change in LVP was observed (P > 0.05). The changes in pressure-rate product reflected the changes in LVP as hearts were driven at a constant pacing rate (data not shown). The end-diastolic pressure was significantly influenced neither by coronary ligation nor by either treatment (P > 0.05) (data not shown). Ventricular fibrillation was not observed and there was no difference in the incidence of arrhythmias between the groups. Coronary increase of epicardial ligation led to a significant (P < 0.05) NADH surface fluorescence that was completely reversible upon reperfusion (Fig 5). There was no significant difference in myocardial ischemia extent between the groups 1

Table 1. Actual Values of the Left Ventricular Pressure and the Global Coronary Flow Before Coronary Occlusion (t = 0 min) and After an lschemic Period of 30 minutes CF (mL/min)

LVP (mmHg) Time

(min)

Control

0

30

0

30

95 + 6

64 + 6

23 -t 2

16 k 2

73 + 5

24 2 2

18 2 2

78 * 8

25 2 3

21 ‘- 3

77 t 7

242

19 r 2

Ouabain

10423

lsoproterenol

103?

Amrinone

107 r4

5

1

NOTE. The values are given as the mean -t SEM, N = 6 in each group. There was no significant difference among the four groups (P > 0.05).

minute or 30 minutes after coronary occlusion before the application of inotropes was started (P > 0.05) (Table 2). Epicardial NADH fluorescence area was not significantly influenced by ouabain or isoproterenol treatment (P > 0.05) (Fig 5). Amrinone significantly diminished NADH fluorescence area (P < 0.05) (Fig 5). However, myocardial ischemia increased again over time to attain almost its initial value before amrinone application at t = 120 minutes (Fig 5).

DISCUSSION

The isolated heart perfused via its coronary circulation (Langendorff preparation) is a well-standardized and extensively used preparation in cardiac pharmacology.20 It permits the study of contractility and coronary dilator activity of drugs. The major advantages include ease of control of temperature and ionic environment as well as heart rate in paced preparations. Moreover, poorly controlled variables such as autonomic nervous system activity and druginduced changes in venous return and peripheral resistance can be negated. Epicardial NADH fluorescence permits visualization of changes of tissue-reduced pyridine nucleotides (primarily NADH), and has been shown to be a very sensitive indicator of myocardial oxygenation.21l23 UV-flash photography of NADH fluorescence 23-2shas the unique advantage of allowing multiple assessment of tissue oxygenation over the experimental time without perturbing heart perfusion.

RUMP ET AL

576

Left Ventticulat Pressure A%

Amrinons

+eo*SO-

-lOt 0

I1 30

60

90

If

I 140

120 nme

Furthermore, a computer-aided analysis of NADH fluorescence pictures allows quantification of myocardial &hernia extent.** One major problem is, however, inherent with this methodology: NAD redox state is measured only in the epicardium. Nevertheless, it was shown that NADH fluorescence patterns on the heart surface are characteristic of the ventricle wall as a whole.26,27 It was demonstrated by comparison of epicardial NADH fluorescence and ATP distribution pattern in myocardial cross-sections that epicardial NADH fluorescence actually reflects transmural ischemia.2s Thus, UV-flash photography is an adequate technique to investigate the effects of pharmacologic agents on acute myocardial ischemia. A%

hnln)

Fig 4. Time course of the relative changes of the left ventricular pressure over the experiment as shown in the experimental protocol in Fii 2. Coronary ligation was set at the time t = 30 minutes and the respective values at t = 30 mimes were used as a reference (160%). Points represent the mean * SEfUl, N = 6 in each group.

Myocardial O2 demand depends on heart rate, wall tension, and contractility.29-31 Because hearts were paced at a constant rate over the course of the experiment, the influence of heart rate on myocardial ischemia extent does not need further consideration in this model. Furthermore, because left ventricular pressure was measured isovolumetrically, a reduction in wall tension by changes in ventricular geometry can also be largely excluded. The main determinant of myocardial 02 consumption in this model is, therefore, contractility and, as one of its indicators, the actively developed left ventricular pressure. The three inotropic agents increased left ventricular pressure and pressure-rate product to a similar extent, also suggesting a

+50

0

-50

i

0 min

30

min

40

min

60

min

90

min

120 min

140

min

Fig 5. Area of the ischemic zone over the experiment as shown in the experimental protocol in Fig 2. The ischemic erea after an ischemic period of 30 minutes was used as a reference (100%). Columns represent the mean f SEM, N = 6 in each group. *Significance at P = 0.06. 0, control; 0, ouabain; E$ isoprenaline; n, amrinone.

EFFECTS OF AMRINONE ON MYOCARDIAL ISCHEMIA

575

GlobalCoronaryFlow I 1

lrchemia Ouabain lsonrenaline

A % ‘60

+so +40

+30 +20

*ia Fig 3. Time course of the relative changes of the global coronary flow over the experiment as shown in the experimental protocol in Fig 2. Coronary ligation was set at the time t = 0 minutes and the respective values at t = 30 minutes were used as a reference (100%). Points represent the mean f SEM, N = 6 in each group.

C -11 -2’ I

I

0

30

in CF by amrinone was, however, significantly higher than after isoproterenol application (P < 0.05) (Fig 3). Whereas CF slightly but significantly decreased again after t = 90 minutes (P < 0.05) no significant change in LVP was observed (P > 0.05). The changes in pressure-rate product reflected the changes in LVP as hearts were driven at a constant pacing rate (data not shown). The end-diastolic pressure was significantly influenced neither by coronary ligation nor by either treatment (P > 0.05) (data not shown). Ventricular fibrillation was not observed and there was no difference in the incidence of arrhythmias between the groups. Coronary ligation led to a significant (P < 0.05) increase of epicardial NADH surface fluorescence that was completely reversible upon reperfusion (Fig 5). There was no significant difference in myocardial ischemia extent between the groups 1

Table 1. Actual Values of the Left Ventricular Pressure and the Global Coronary Flow Before Coronary Occlusion (t = 0 min) and After an lschemic Period of 30 minutes CF imL/minl

LVP(mmHgj 0

Ii0

kz-70 llnw

increase

Time (min)

I

60

30

0

30

Control

95 k 6

64 k 6

23 k 2

16k

Ouabain

104 k 3

73 r 5

24 k 2

18 k 2

lsoproterenol

103 lr 5

78 2 8

25 + 3

21 k3

Amrinone

107k4

77 * 7

242

19 2 2

1

2

NOTE. The values are given as the mean k SEM, N = 6 in each group. There was no significant difference among the four groups (P > 0.05).

(mlnl

minute or 30 minutes after coronary occlusion before the application of inotropes was started (P > 0.05) (Table 2). Epicardial NADH fluorescence area was not significantly influenced by ouabain or isoproterenol treatment (P :> 0.05) (Fig 5). Amrinone significantly diminished NADH fluorescence area (P < 0.05) (Fig 5). However, myocardial ischemia increased again over time to attain almost its initial value before amrinone application at t = 120 minutes (Fig 5).

DISCUSSION

The isolated heart perfused via its coronary circulation (Langendorlf preparation) is a well-standardized and extensively used preparation in cardiac pharmacology.z0 It permits the study of contractility and coronary dilator activity of drugs. The major advantages include ease of control of temperature and ionic environment as well as heart rate in paced preparations. Moreover, poorly controlled variables such as autonomic nervous system activity and druginduced changes in venous return and peripheral resistance can be negated. Epicardial NADH fluorescence permits visualization of changes of tissue-reduced pyridine nucleotides (primarily NADH), and has been shown to be a very sensitive indicator of myocardial oxygenation.21~23 UV-flash photography of NADH fluorescence23-25 has the unique advantage of allowing multiple assessment of tissue oxygenation over the experimental time without perturbing heart perfusion.

578 consumption in acute experimental cardiac depression. Circ Res 21:123-138,1967 7. Clancy RL, Graham TP Jr, Powell WJ Jr, Gilmore JP: Inotropic augmentation of myocardial oxygen consumption. Am J Physiol212:1055-1061,1967 8. Lesch M: Inotropic agents and infarct size. Am J Cardiol 37508-513, 1976 9. Vatner SF, Baig H: The effects of inotropic stimulation on ischemic myocardium in conscious dogs. Trans Assoc Am Physicians 91:282-293, 1978 10. Storstein L: Digitalis in acute myocardial infarction-risk and benefit. Progress Pharmacol4:35-41,198l 11. Benotti JR, Grossman W, Braunwald E, Carabello BA: Effects of amrinone on myocardial energy metabolism and hemodynamics in patients with severe congestive heart failure due to coronary artery disease. Circulation 62:28-34,198O 12. Holubarsch C, Hasenfuss G, Heiss WH, Just H: Influence of enoximone and UDCG-115 on coronary hemodynamics in idiopathic dilated cardiomyopathy. Am J Cardiol62:104E-107E, 1988 13. Nanto S, Kodama K, Kimura Y, et al: Effect of OPC8212(2(1H)-quinolinone), a new inotropic agent, on myocardial energy metabolism in patients with coronary heart disease. Jpn Circ J 52:155-161, 1988 14. Hasenfuss G, Holubarsch C, Heiss HW, et al: Myocardial energetics in patients with dilated cardiomyopathy. Influence of nitroprusside and enoximone. Circulation 80:51-64, 1989 15. Jentzer JH, Leje’mtel TH, Sonnenblick EH, Kirk ES: Beneficial effect of amrinone on myocardial oxygen consumption during acute left ventricular failure in dogs. Am J Cardiol48:75-83,198l 16. Rude RE, Kloner RA, Maroko PR, et al: Effects of amrinone on experimental acute myocardial ischaemic injury. Cardiovasc Res 14:419-427, 1980 17. Langendorff 0: Untersuchungen am iiberlebenden Siugetierherzen. Arch Ges Physiol61:291-332,1895 18. Roesen R, Panzner B, Klaus W: Epicardial image analysis using a desk top computer. Adv Myocardiol6:217-231,1985 19. Goenen M: Historical perspectives and update of amrinone. J Cardiothorac Anesth 3:15-23,1989 20. Katzung B: Evaluation of drugs affecting the contractility and the electrical properties of the heart, in Burger A (ed): Selected Pharmacological Testing Methods. New York, NY, Marcel Dekker, 1968, pp 193-234 21. Williamson JR: Glycolytic control mechanisms. II. Kinetics of intermediate changes during aerobic-anoxic transition in the perfused rat heart. J Biol Chem 241:5026-5036, 1966 22. Chance B, Salkovitz JA, Kovach AGB: Kinetiks of mitochondrial flavoprotein and pyridine nucleotide in perfused heart. Am J Physiol223:207-218, 1972 23. Barlow CH, Harken AD, Chance B: Evaluation of cardiac

RUMP ET AL

ischemia by NADH fluorescence photography. Ann Surg 186:737740,1977 24. Steenbergen C, Deleeuw G, Williamson JR: Analysis of control of glycolysis in ischemic hearts having heterogeneous zones of anoxia. J Mol Cell Cardiol 10:617-639, 1978 25. Simson MB, Harden W, Barlow C, Harken AH: Visualization of the distance between perfusion and anoxia along an ischemic border. Circulation 60:1151-1155, 1979 26. Harken AH, Barlow CH, Harden WR, Chance B: Two and three dimensional display of myocardial ischemic “border zones” in dogs. Am J Cardiol42:954-959, 1978 27. Williamson JR, Davis KN, Medina-Ramirez G: Quantitative analysis of heterogenous NADH fluorescence in perfused rat hearts during hypoxia and ischemia. J Mol Cell Cardiol 14:29-35, 1982 28. Klaus W, Horneff G, Roesen R: The influence of nifedipine on three dimensional distribution of ATP in regional ischemic rabbit hearts. J Mol Cell Cardiol21:152, 1989 29. Braunwald E.: Control of myocardial oxygen consumption. Am J Cardiol27:416-432,197l 30. Baller D, Bretschneider HJ, Hellige G: Validity of myocardial oxygen consumption parameters. Clin Cardiol 2:317-327, 1979 31. Baller D, Wolpers HG, Hoeft A, et al: Increase of myocardial oxygen consumption due to active diastolic wall tension. Basic Res Cardiol76:176-185, 1984 32. Su YE, Harden TK, Perkin JP: Isoproterenol-induced desensitazition of adenylate cyclase in human astrocytoma cells. J Biol Chem 254:38-41, 1979 33. Schrader J, Baumann G, Gerlach E: Adenosine as inhibitor of myocardial effeats of catecholamines. Pfliigers Arch 372:29-35, 1977 34. Schaper W: Regulation of coronary blood flow, in Schaper W (ed): The pathophysiology of myocardial perfusion. New York, NY, ElsevieriNorth-Holland Biomedical, 1979 35. Fujino K, Sperelakis N, Solar0 RJ: Sensitization of dog and guinea pig heart myofilaments to Ca2+ activation and the inotropic effect of pimobendan:comparison with milrinone. Circ Res 63:911922,1988 36. Cohen MV, Kirk ES: Differential response of large and small coronary arteries to nitroglycerin and angiotensin. Circ Res 33:445-453,1973 37. Gage J, Rutman H, Lucid0 D, LeJemtel H: Additive effects of dobutamine and amrinone on myocardial contractility and ventricular performance in patients with severe heart failure. Circulation 74:367-373, 1986 38. Robinson RJS, Tchervenkov C: Treatment of low cardiac output after aortocoronary artery bypass surgery using a comhination of norepinephrine and amrinone. J Cardiothorac Anesth 3:229-233, 1987