Amrinone is a bipyridine compound that inhibits the low Km cyclic adenosine monophosphate (cAMP)-specific, cyclic guanosine monophosphate -inhibited phosphodiesterase enzyme. The increased intracellular cAMP levels that result from phosphodiesterase inhibition can produce positive inotropic and vasodilatory effects via cAMP-mediated changes in calcium flux in myocytes and vascular smooth muscle cells [1-3]. Amrinone increases cardiac index, decreases systemic vascular resistance, lowers filling pressures, and is effective in adult patients for the treatment of congestive heart failure and low cardiac output syndrome after cardiac surgery [4,5]. While the experience with amrinone in pediatric patients is more limited, several studies have demonstrated that it increases cardiac output while decreasing systemic and pulmonary resistance [6-9]. However, it is not clear whether the improvement in cardiac output is primarily due to inotropic or vasodilatory effects. This question is of particular interest in neonates and infants, since amrinone has a negative inotrope in newborn canine and neonatal porcine myocardium [10,11], suggesting the possibility that increases in cardiac output are due to vasodilation. If amrinone increases cardiac output in infants primarily via vasodilation, other vasodilators, such as sodium nitroprusside, might be more rational for the treatment of low cardiac output syndrome since they are less expensive and, more importantly, the pharmacokinetic profile of amrinone is not conducive to rapid titration of plasma levels [12,13]. The purpose of this study was to address the relative contributions of amrinone's inotropic action and vasodilation in infants indirectly by comparing the hemodynamic effects of amrinone (AMR) and sodium nitroprusside (SNP).
The study protocol was approved by the Human Investigations Committee of Emory University School of Medicine. Informed consent was obtained from the parents or guardians of 10 pediatric patients, aged less than or equal to 12 mo, scheduled for elective cardiac surgery. Patient demographics are presented in Table 1, including ischemic time (duration of aortic cross-clamping) and reperfusion time (the interval from release of the aortic cross-clamp to commencing the study).
Patients were studied after surgical repair of the presenting defect, separation from cardiopulmonary bypass (CPB), and administration of protamine. Transesophageal echocardiography (TEE) was used to assess the surgical repair in eight patients (equipment was not available for two patients), and all patients had blood samples taken from the superior vena cava, right atrium, and pulmonary artery for hemoglobin oxygen saturation measurement. No patient in the study had a residual intracardiac shunt by either of these criteria. Prior to administration of either SNP or AMR, a 2.5 Fr thermistor was placed in the main pulmonary artery by the surgical team. Cardiac output was measured by thermodilution with the injection of 3 or 5 mL of cold (4 degrees C) saline into the distal port of a double-lumen central venous catheter placed via the right internal jugular vein with the tip at the superior vena caval-right atrial junction. All patients were hemodynamically stable with inotropic support at the time of the study (Table 1).
The study began with the recording of control hemodynamic variables (heart rate, systemic blood pressure, left atrial pressure, central venous pressure, and cardiac output [in triplicate]) and midpapillary shortaxis images of the left ventricle (in eight patients). SNP was then titrated to decrease mean blood pressure (MBP) by 20% or to a value of [approximately]50 mm Hg. Hemodynamic variables and echocardiographic images were again recorded after MBP had stabilized (defined as less than a 10% change over 1 min). SNP was then discontinued and, after MBP had stabilized, a second set of control hemodynamic and TEE measures were recorded. Then, 1.5 mg/kg of AMR was administered as single bolus dose and, after stabilization of MBP, hemodynamic and TEE measures were made.
Throughout the titration of SNP, its discontinuation, and administration of AMR, left atrial pressure was maintained at the level used for separation from CPB by the administration of an autologous blood/albumin (50:50) mixture.
End-diastolic areas were calculated from short-axis views, when available, of the left ventricle using the leading edge-leading edge convention without inclusion of the papillary muscles .
Repeated measures analysis of variance was used to determine whether significant changes occurred in hemodynamic variables during the study. For those variables with significant changes, paired t-tests with the Bonferroni correction were used to isolate pairwise differences. The relationship between fractional change in cardiac index (CI) after SNP or AMR was investigated in a limited fashion using linear regression as a function of initial CI, initial systemic vascular resistance (SVRI), initial MBP, MBP after SNP or AMR, ischemic time, and reperfusion time. Statistical analysis was implemented using Systat[TM] (Evanston, IL).
The mean ischemic time for patients enrolled in this study was 44 min with a standard deviation of 12 min. The mean reperfusion time was 52 min with a standard deviation of 14 min. The study was begun 21 min (mean) +/- 6 min (SD) after separation from CPB. The study was completed within 40 min in all patients, requiring no more than 20 min for titration of SNP, no more than 10 min of stabilization for the second control measurements, and no more than 5 min for stabilization after AMR.
Repeated measures analysis of variance indicated significant changes in MBP (P = 4.1 x 10-6), CI (P = 4.8 x 10-6), and SVRI (P = 6.7 x 10-7) during the study (Table 2). By paired t-test, MBP and SVRI decreased, in comparison with control values, after both SNP and AMR. CI increased significantly after AMR but not after SNP (Table 2). There were no differences in any of the variables between the two control measurements.
The ratio of fractional change in CI to absolute fractional change in MBP was significantly higher for AMR than SNP (Figure 1).
Because of the unexpected observation that CI did not significantly increase after SNP, the relationships between fractional change in CI and ischemic time, reperfusion time, initial CI, MBP, and SVRI, and MBP after SNP were analyzed by linear regression. There was a significant correlation between fractional change in CI after SNP and initial MBP (r = 0.74, P = 0.015), MBP after SNP (r = 0.77, P = 0.0089), and initial SVRI (r = 0.66, P = 0.039). Other correlations were not significant.
Catecholamines are commonly used drugs for the treatment of low cardiac output syndrome after cardiac surgery. These drugs produce a positive inotropic effect by stimulating the synthesis of cAMP, a major regulator of intracellular calcium. The disadvantages of catecholamines are increased heart rate and increased afterload secondary to vasoconstriction from alpha-adrenergic effects. Furthermore, adult patients in preoperative heart failure may have down-regulated beta-adrenergic receptors . AMR and other phosphodiesterase inhibitors offer an alternate therapy that potentially avoids the disadvantages of catecholamines. Several studies have demonstrated improved myocardial function after AMR administration to infants, children, and adults [5-9,16]. However, since AMR in vitro has both inotropic and vasodilatory effects, it is unclear which, if either, has the dominant effect on pump performance. Several early studies in adults suggested that AMR increased cardiac output primarily by vasodilation [17-19]. A subsequent investigation comparing AMR with the pure vasodilator SNP indicated that in adults AMR increases cardiac output both by vasodilation and via a positive inotropic effect . However, this issue had not yet been addressed for pediatric patients, and it seemed a particularly significant relevant question since early studies had shown that AMR had a negative inotropic effect in neonatal dogs and pigs. Clearly, if the beneficial effects of AMR are due primarily to vasodilation, other drugs would be more rational than AMR since the pharmacokinetics of AMR are not conducive to rapid titration of drug effect.
The most direct and elegant approach to evaluate the relative inotropic effect of AMR would be the assessment of a load-independent measure of contractility such as end-systolic elastance, preload recruitable stroke work, or the left ventricular end-systolic wall stress velocity of fiber shortening index [21-23]. This is technically difficult in the postoperative period, and the measurement of ventricular dimension can be especially challenging in infants. We attempted to consider this issue indirectly by comparing the hemodynamic effects of AMR and SNP while maintaining preload at a constant level and with each patient serving as his or her own control. Our basic question was, if other determinants of cardiac output (heart rate and preload) are constant, which drug produced the largest increase in cardiac output per unit decrease in blood pressure? A large difference in the increase in cardiac output/decrease in blood pressure ratio may indicate a significant inotropic effect. Since the most common limitation to the use of vasodilators for the treatment of low cardiac output is hypotension, we view this ratio as a practical measure of the clinical utility of these drugs.
Our primary finding was that AMR increased cardiac output significantly more than SNP and that the ratio of fractional increase in cardiac output to fractional absolute decrease in MBP was significantly greater with AMR. Heart rate and preload, measured by left atrial pressure and by left ventricular cross-sectional area (in eight patients), were constant throughout the protocol. Our interpretation of these findings is that AMR increases cardiac output via a significant positive inotropic effect in infants after cardiac surgery since with constant preload and heart rate AMR increases cardiac output more per unit decrease in afterload than a pure vasodilator, SNP.
Our interpretation of these results is subject to several caveats. It was necessary to study AMR after the effects of SNP were analyzed, since AMR has a much longer effective half-time than SNP. It is possible that underlying increases in myocardial contractility, related to recovery from the effects of cardiac surgery and independent of either drug, led to an apparent therapeutic superiority of AMR, since it was the last drug studied. We do not think this is realistic, however, since the differences in control measurements before and after SNP were not statistically significant and since the time interval from the second set of control measurements and completion of the study was short (<10 min). Also we did not find a significant correlation between fractional change in CI and either ischemic time or reperfusion time.
Another major concern is our unanticipated observation that CI did not increase significantly after SNP. Previous investigations of pediatric patients have shown SNP to be an effective therapy for low cardiac output syndrome . While we cannot fully explain this discrepancy, we did note a significant correlation between fractional change in CI after SNP and initial MBP, initial SVRI, and MBP after SNP. We postulate that SNP was ineffective in this study because not all patients were vasoconstrictive or hypertensive. Several patients had a MBP less than 50 mm Hg after SNP. Animal studies clearly demonstrate that the effect of vasodilation on CI is limited by blood pressure . However, we also cannot rule out subtle differences in preload that could not be detected by changes in left atrial pressure or by limited echocardiography. This study was conducted after protamine neutralization of heparin but prior to administration of procoagulants such as platelets. This is a period characterized by fluid shifts, due not only to frank bleeding but also to the third-space losses attributable to capillary leak and to brisk diuresis secondary to mannitol and furosemide administered during CPB. Despite volume loading during SNP administration, left atrial pressure and central venous pressure remained constant after it was discontinued during the second control period, presumably due to ongoing loss of volume from the intravascular space. If the preload was decreased after SNP, CI might actually decrease. Furthermore, each patient required additional intravascular fluid administration after AMR (up to 10 mL/kg), and the positive effect of AMR on CI could simply reflect these additional fluids. However, we emphasize that the measures of preload available to us remained constant during the study despite these fluid shifts. Any differences in preload were too subtle to detect by conventional measures.
In the same vein, we also note that phosphodiesterase inhibitors improve diastolic function . We cannot rule out the possibility that the increase in cardiac output produced by AMR was due in part to improved diastolic function with a higher effective end-diastolic volume at the same left atrial pressure and not detected by our limited echocardiography.
A final caveat is that most of the patients in this study had an adequate CI at the onset of the study. We might have found a greater effect of SNP on CI if we had studied only patients with low cardiac output syndrome. However, the logistics of this were prohibitive.
In summary, several authors have commented on the relative advantages that AMR offers in comparison with catecholamines for the treatment of low cardiac syndrome. In this study, we have shown that AMR also offers advantages over a commonly used vasodilator, SNP, increasing cardiac output more per unit decrease in blood pressure. The best explanation for this observation is that AMR has a positive inotropic effect in infants after cardiac surgery that is clinically significant.
1. Alousi A, Farah A, Lesher G, et al. Cardiotonic activity of amrinone-WIN 40680. Circ Res 1979;45:666-77.
2. Weishaar RE, Burrows SD, Kobylarz DC, et al. Multiple molecular forms of cyclic nucleotide phosphodiesterase in cardiac and smooth muscle and platelets: isolation, characterization, and effects of various reference phosphodiesterase inhibitors and cardiotonic agents. Biochem Pharmacol 1986;35:787-800.
3. Benotti JR, Grossman W, Braunwald E, et al. Hemodynamic assessment of amrinone. N Engl J Med 1987;299:1373-7.
4. Mancini D, Lejemtel TH, Sonnenblick EH. Intravenous use of amrinone for the treatment of the failing heart. Am J Cardiol 1985;56:8B-15B.
5. Butterworth JF, Royster RL, Prielipp RC, et al. Amrinone in cardiac surgical patients with left-ventricular dysfunction. Chest 1993;104:1660-7.
6. Lawless S, Burkhart G, Diven W, et al. Amrinone in neonates and infants after cardiac surgery. Crit Care Med 1989;17:751-4.
7. Robinson BW, Gelbard H, Mas MS. Selective pulmonary and systemic vasodilator effects of amrinone in children: new therapeutic implications. J Am Coll Cardiol 1993;21:1461-5.
8. Wessel DL, Triedman JK, Wernovsky G, et al. Pulmonary and systemic hemodynamic effects of amrinone in neonates following cardiopulmonary bypass. Circulation 1989;80(Suppl II):II-488.
9. Lynn AM, Sorenson GK, Williams GD, et al. Hemodynamic effects of amrinone and colloid administration in children following cardiac surgery. J Cardiothorac Vasc Anesth 1993;7:560-5.
10. Binah O, Banilo P, Rosen MR, et al. Developmental changes in the cardiac effects of amrinone in the dog. Circ Res 1983;52:747-53.
11. Ross-Ascuitto N, Ascuitto R, Chen V, et al. Negative inotropic effects of amrinone in the neonatal piglet heart. Circ Res 1987;61:847-52.
12. Lawless S, Burkhart G, Diven W, et al. Amrinone pharmacokinetics in neonates and infants. J Clin Pharmacol 1988;28:283-4.
13. Bailey JM, Levy JH, Rogers HG, et al. Pharmacokinetics of amrinone during cardiac surgery. Anesthesiology 1991;75:961-8.
14. Wyatt HL, Haendchen RV, Meerbaum S, et al. Assessment of quantitative methods of 2-dimensional echocardiography. Am J Cardiol 1983;52:396-401.
15. Bristow MR, Ginsburg R, Minobe W, et al. Decreased catecholamine sensitivity and beta-adrenergic receptor density in failing human hearts. N Engl J Med 1982;307:205-11.
16. Levy JH, Bailey JM. Amrinone: pharmacokinetics and pharmacodynamics. J Cardiothorac Anesth 1989;80(Suppl 2):10-4.
17. Firth BG, Ratner AV, Grassman ED, et al. Assessment of the inotropic and vasodilator effects of amrinone versus isoproterenol. Am J Cardiol 1984;54:1331-6.
18. Hermiller JB, Leithe ME, Magorien RD, et al. Amrinone in severe congestive heart failure: another look at an intriguing new cardioactive drug. J Pharmacol Exp Ther 1984;228:319-26.
19. Wilmshurst PT, Thompson DS, Jenkins BS, et al. Effects of intracoronary and intravenous amrinone infusions in patients with cardiac failure and patients with near normal cardiac function. Br Heart J 1985;53:493-506.
20. Konstam MA, Cohen SR, Weiland DS, et al. Relative contribution of inotropic and vasodilator effects to amrinone-induced hemodynamic improvement in congestive heart failure. Am J Cardiol 1986;57:242-8.
21. Sagawa K. The end-systolic pressure-volume relation of the ventricle: definition, modifications, and clinical use. Circulation 1981;63:1223-7.
22. Glower DD, Spratt JA, Snow ND, et al. Linearity of the Frank-Starling relationship in the intact heart: the concept of preload recruitable stroke work. Circulation 1985;48:813-24.
23. Colan SD, Borow KM, Neumann A. The left ventricular end-systolic wall stress-velocity of fiber-shortening relation: a load-independent index of myocardial contractility. J Am Coll Cardiol 1984;4:715-23.
24. Applebaum A, Blackstone E, Kouchoukos NT, et al. Afterload reduction and cardiac output in infants after intracardiac surgery. Am J Cardiol 1977;39:445-51.
25. Isoyama S, Maruyama Y, Koiwa Y, et al. Experimental study of afterload-reducing therapy. The effects of the reduction of systemic vascular resistance on cardiac output, aortic pressure, and coronary circulation in isolated, ejecting canine hearts. Circulation 1981;64:490-9.
26. Monrad ES, McKay RG, Baim DS, et al. Improvement in indexes of diastolic performance in patients with congestive heart failure treated with milrinone. Circulation 1984;70:1030-7.