Calcium requirement for cGMP production during muscarinic activation of N1E-115 neuroblastoma cells.

Muscarinic agonists elicit large increases in intracellular Ca2+ and guanosine 3',5'-cyclic monophosphate (cGMP) in N1E-115 neuroblastoma cells. Both signals are blocked in cells loaded with the Ca2+ buffer 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid showing that the increase in intracellular Ca2+ concentration ([Ca2+]i) is necessary to stimulate cGMP accumulation. Inhibition of nitric oxide synthase (NOS) blocks the cGMP response without affecting the peak amplitude of the intracellular Ca2+ signal, and it is concluded that Ca(2+)-dependent activation of NOS is required for cGMP production. cGMP accumulation is reduced by 60% when cells are bathed in Ca(2+)-free saline, but the peak change in [Ca2+]i is not affected. This suggests that Ca2+ influx is strongly coupled to the activation of cGMP production, even though it makes a smaller contribution to the intracellular Ca2+ signal than does Ca2+ release. Thapsigargin, which releases Ca2+ from intracellular stores, activates Ca2+ influx and increases cGMP. The cGMP increase is transient and follows approximately the same time course as Ca2+ store depletion. Ca2+ influx remains activated after store depletion, however, which indicates that influx alone cannot sustain cGMP production. It is concluded that summation of Ca2+ influx and Ca2+ release is necessary to reach a threshold Ca2+ level needed to stimulate cGMP accumulation. Because of the large contribution from Ca2+ influx, we suggest that NOS or a cofactor necessary for its activation may be located close to Ca2+ channels in the membrane.

The lifetime of inositol 1,4,5-trisphosphate in single cells.

In many eukaryotic cell types, receptor activation leads to the formation of inositol 1,4,5-trisphosphate (IP3) which causes calcium ions (Ca) to be released from internal stores. Ca release was observed in response to the muscarinic agonist carbachol by fura-2 imaging of N1E-115 neuroblastoma cells. Ca release followed receptor activation after a latency of 0.4 to 20 s. Latency was not caused by Ca feedback on IP3 receptors, but rather by IP3 accumulation to a threshold for release. The dependence of latency on carbachol dose was fitted to a model in which IP3 synthesis and degradation compete, resulting in gradual accumulation to a threshold level at which Ca release becomes regenerative. This analysis gave degradation rate constants of IP3 in single cells ranging from 0 to 0.284 s-1 (0.058 +/- 0.067 s-1 SD, 53 cells) and a mean IP3 lifetime of 9.2 +/- 2.2 s. IP3 degradation was also measured directly with biochemical methods. This gave a half life of 9 +/- 2 s. The rate of IP3 degradation sets the time frame over which IP3 accumulations are integrated as input signals. IP3 levels are also filtered over time, and on average, large-amplitude oscillations in IP3 in these cells cannot occur with period < 10 s.

Alteration in Gs-mediated signal transduction in S49 lymphoma cells treated with inhibitors of microtubules.

We have assessed the possible interaction between the microtubular component of the cytoskeleton and signal transducing GTP-binding (G) proteins by examining the ability of colchicine and vinblastine (two microtubule disrupters) to alter Gs and Gi protein activity in S49 lymphoma cells. Treatment of wild type S49 cells with cholchicine and vinblastine increased beta-adrenergic agonist- and prostaglandin (PG) E1-stimulated formation of cAMP. The microtubular inhibitor nocodazole also enhanced isoproterenol-stimulated cAMP accumulation, whereas the inactive analog of colchicine, beta-lumicolchicine, did not have this action. Based on data obtained with wild type, cyc-, and UNC S49 cells, we determined that enhancement in cyclic AMP accumulation is proximal to the catalytic (C) unit of adenylylcyclase, distal to hormone receptors, and seems to be located on Gs. Treatment with colchicine increased guanosine 5'-(gamma-thio)triphosphate-stimulated accumulation of cAMP in permeabilized wild type cells. The increase in activity of Gs appeared not to result from a change in the intracellular concentration of GTP. Treatment of cells with colchicine or vinblastine also increased the amount of the alpha s-C complex, as assessed by the binding of [3H]forskolin to intact cells at 37 degrees C. In contrast to the observed effect on Gs, treatment of wild type S49 cells with colchicine failed to modify the degree of inhibition of cAMP formation produced by somatostatin, which acts via the activation of Gi. These data suggest that microtubules regulate the ability of Gs to interact with and activate the catalyst of adenylylcyclase.

Stoichiometry of receptor-Gs-adenylate cyclase interactions.

Little is known about the relative stoichiometry of guanine nucleotide-binding (G) proteins relative to the effector systems to which they link. We addressed this question for the stimulatory G protein (Gs) linked to adenylate cyclase. Forskolin stimulates the catalytic subunit of adenylate cyclase (C), but it has a higher efficacy and potency when C also interacts with the G protein Gs. Accordingly, we measured high-affinity [3H]forskolin binding to intact cells to assay alpha s-C complexes. No high-affinity specific binding occurred with unstimulated cells. The beta-adrenergic agonist isoproterenol promoted the binding of [3H]forskolin to about 3000 sites per cell, suggesting that each receptor on average activates at least several Gs molecules. Activating Gs directly with cholera toxin maximally promoted [3H]forskolin binding to a similar number of sites, suggesting that this is the maximal number of alpha s-C complexes formed per cell. We conclude that each cell likely contains only a few thousand functional copies of C, and that the availability of C (rather than Gs, which exists in more than 100,000 copies per cell) is likely to be limiting for agonist stimulation of adenylate cyclase activity.

Reduced cardiac myofibrillar Mg-ATPase activity without changes in myosin isozymes in patients with end-stage heart failure.

In this study we tested the hypothesis that reduced myofibrillar ATPase activities in end-stage heart failure are associated with a redistribution of myosin isozymes. Cardiac myofibrils were isolated from left ventricular free wall from normal human hearts and hearts at end-stage heart failure caused by coronary artery diseases, cardiomyopathy or immunological rejection. The hearts had been excised in preparation for a heart transplant. Myofibrillar Ca2(+)-dependent Mg-ATPase and myosin Ca2(+)- and K+EDTA-ATPase activities were compared. Possible changes in myosin isozyme distribution in the diseased heart were investigated using polyacrylamide gel electrophoresis of native myosin in the presence of pyrophosphate. Significant reduction in myofibrillar Ca2(+)-dependent Mg-ATPase with no changes in the sensitivity of the myofibrils to Ca2+ was observed in heart with coronary artery diseases (25.2 to 27.1% at pCa 5.83 to pCa 5.05), cardiomyopathy (21.1 to 25.5% at pCa 5.41 to pCa 5.05), and in the immunologically rejected heart (18.4 to 22.8% at pCa 5.41 to pCa 5.05). Significantly lower myosin Ca2(+)-ATPase was observed with coronary artery diseases only and myosin K-EDTA activities did not differ in diseased and normal hearts. Polyacrylamide gel electrophoresis of native myosin from the normal and three models of end-stage heart failure revealed two distinct bands in the human left ventricle and one diffuse band in the human right atria. No apparent differences in myosin isoenzyme pattern were observed between the normal and diseased hearts. Further evaluation is needed to clarify the ATPase nature of the two bands.

Differential vasorelaxant effects of milrinone and amrinone on contractile responses of canine coronary, cerebral, and renal arteries.

The vasorelaxant effects of milrinone and amrinone in canine coronary, cerebral, and renal arterial rings or strips contracted by either K+-depolarization, U46619 (a thromboxane mimetic), or prostaglandin F2 alpha (PGF 2 alpha) were quantitated. Milrinone was more potent as a vasorelaxant in coronary arteries relative to cerebral or renal arteries regardless of the mode of contraction; amrinone was coronary selective with K+ contraction only. When comparing potency in arteries contracted by different agonists, milrinone was significantly more potent as a vasorelaxant in all three arteries contracted by either U46619 or PGF2 alpha than in arteries contracted by K+ depolarization, whereas amrinone was only selective for U46619-induced contractions in cerebral arteries. This profile of activity for milrinone was similar to that of sodium nitrite and isoproterenol and dissimilar from the calcium entry blocking agents nimodipine and nifedipine. In conclusion, this study shows that coronary vascular selectivity exists for milrinone and amrinone. Moreover, the relaxant profiles of milrinone and amrinone, with different sources of vascular smooth muscle, are unlike those of calcium entry blocking agents and more similar to the profiles of agents that modulate cyclic nucleotide levels.

Changes in myofibrillar content and Mg-ATPase activity in ventricular tissues from patients with heart failure caused by coronary artery disease, cardiomyopathy, or mitral valve insufficiency.

Force development and shortening by cardiac muscle occur as a result of the interaction between actin and myosin within the myofibrillar lattice. This interaction is dependent upon intracellular ionized calcium and is controlled by the troponin-tropomyosin regulatory proteins situated along the actin filament. In this study, we compared the myofibrillar content and myofibrillar Mg-ATPase activity of normal human ventricular muscle with that of ventricular muscle from patients in end-stage failure caused by coronary artery disease or cardiomyopathy and ventricular muscle from patients with heart failure due to mitral valve insufficiency. The results show that the amount of myofibrillar protein (mg/g wet wt ventricle) in hearts in end-stage failure (coronary artery disease and cardiomyopathy) is significantly lower compared with normal hearts and hearts in failure due to mitral valve insufficiency. However, the Mg-ATPase activity of myofibrils from hearts in both end-stage failure and failure due to mitral valve insufficiency is significantly lower compared with myofibrils from normal hearts. The data suggest that the reduction in the amount of myofibrillar protein in ventricular tissue is a pivotal event that may be responsible for the progression of heart disease to the point of end-stage failure.

Effects of milrinone on Ca++-sensitivity of myofibrillar Mg-adenosine triphosphatase isolated from normal human and canine hearts.

Sensitization or desensitization of cardiac actomyosin to calcium has been demonstrated with several pharmacological agents. The effect of milrinone on the sensitivity of cardiac Mg-adenosine triphosphatase (ATPase) activity to calcium was studied in purified myofibrils isolated from normal human hearts (after accidental death or trauma that caused no cardiac damage as established by the attending physician) and from normal canine hearts (established by echocardiography), over a range of calcium concentrations (pCa, 8 to 5). Caffeine, a cardiac stimulant that has been shown to increase the sensitivity of myofibrillar Mg-ATPase activity to calcium in rat ventricle, was used in this study to establish its effect on canine and human myofibrils in comparison with that of milrinone. Caffeine, at concentrations of 40 mM, caused statistically significant sensitization of canine and human myofibrils to calcium. In canine myofibrils, the calcium-dependent Mg-ATPase activity increased from 11.0 +/- 1.2 to 18.8 +/- 2.6 nmol of Pi per mg of protein per min at pCa 6.73 (N = 9, P less than .05) and from 32.9 +/- 2.1 to 37.3 +/- 2.2 nmol of Pi per mg of protein per min at pCa 6.16 (N = 9, P less than .05), whereas total Mg-ATPase activity increased from 23.4 +/- 1.5 to 33.6 +/- 2.6 nmol of Pi per mg of protein per min at pCa 6.73 (N = 9, P less than .05) and from 45.2 +/- 2.2 to 52.2 +/- 2.5 nmol of Pi per mg of protein per min at pCa 6.16 (N = 9, P less than .05).(ABSTRACT TRUNCATED AT 250 WORDS)

Pharmacology of the bipyridines: amrinone and milrinone.

The cardiovascular bipyridines amrinone and milrinone are positive inotropic agents with vasodilator properties. When their effects on the heart and circulation were studied in both isolated cardiac tissues and anesthetized animals, milrinone was found to be 30 times more potent than amrinone. The inotropic response of isolated atria and papillary muscles to the bipyridines was characterized by increases in isometric tension and in the rates of tension development and relaxation. In the anesthetized dog, amrinone (1.0 mg/kg iv) and milrinone (0.03 mg/kg iv) caused significant increases in left ventricular dP/dtmax, cardiac output, and heart rate, significant decreases in pulmonary arterial pressure, and no change in systemic blood pressure. Studies to elucidate the mechanism(s) of action of the bipyridines indicated that Ca++ and cyclic AMP (cAMP) play important roles. The inotropic response to milrinone was modified by changes in the rate of stimulation and the concentration of extracellular Ca++ and by depolarization with high K+ or pretreatment with a calcium-channel blocker. Reduction of the extracellular Na+ concentration of 75% of normal potentiated the inotropic response of papillary muscles to low concentrations of milrinone, suggesting possible activation of the Na+-Ca++ exchange mechanism. These results suggest that milrinone may stimulate the influx of Ca++ into the cardiac cell. The bipyridines are phosphodiesterase inhibitors that increase cardiac cAMP levels. However, a time-course analysis of the changes in cAMP levels during the inotropic response to the bipyridines indicated that the increase in isometric tension development preceded the increase in cAMP. Our data suggest that more than one mechanism may be involved in the initiation and maintenance of the inotropic response to the bipyridines.

The beneficial effect of amrinone on acute drug-induced heart failure in the anaesthetised dog.

Amrinone, a positive inotropic-vasodilator agent, was administered to anaesthetised dogs in an attempt to reverse heart failure induced by drugs possessing negative inotropic properties. Propranolol, a beta-adrenergic blocker; verapamil, a calcium slow-channel blocker procainamide, a type 1 antiarrhythmic agent; or sodium pentobarbital, a barbituate; administered as a bolus injection and/or infusion, produced a sustained depression in canine cardiac function. Cardiac depression was characterised by a greater than 40% reduction in cardiac contractile force (CF) and maximum left ventricular pressure development (LV dp/dtmax), a 30 to 50% reduction in cardiac output (CO) and concomitant increases in mean central venous or mean right atrial blood pressures (CVP, RAP, respectively). Amrinone, when administered intravenously as a bolus injection (1 or 3 mg X kg-1) plus an infusion (0.03 or 0.1 mg X kg-1 X min-1) reversed the depression in cardiac function by increasing CF, CO and LV dp/dtmax and decreasing preload CVP or RAP in all four drug-induced failure models. Due to the vasodilator properties of amrinone, afterload, total peripheral resistance (TPR), was reduced in verapamil and procainamide failures as well as in propranolol failure, the only model where TPR increases. In another model of heart failure, in which ouabain-induced arrhythmias preceded procainamide toxicity, amrinone was also an effective cardiotonic agent. Ouabain's inotropic effect was studied in propranolol-induced heart failure. Although an increase in LV dp/dtmax and a decrease in CVP were noted, ouabain (40 micrograms X kg-1 iv) increased TPR and had little effect on the depression in CF and CO. Drug-induced models of heart failure were useful pharmacological tools for evaluating the cardiotonic agent's ability to overcome severe cardiac depression. In propranolol-, verapamil-, procainamide-, and pentobarbital-induced cardiac toxicity, amrinone could be of therapeutic value.

Correlation of the hemodynamic and pharmacokinetic profile of intravenous milrinone in the anesthetized dog.

The relationship between the hemodynamic effects of milrinone and its plasma concentration was studied in the anesthetized instrumented dog. Milrinone was administered intravenously either as a single bolus of 10, 30 or 100 micrograms/kg or infused at a rate of 10 micrograms/kg/min. The changes in drug plasma concentration and cardiovascular parameters were determined simultaneously during the course of drug action. The intravenous bolus injections of milrinone caused dose-dependent increases in its maximum plasma concentration that resulted in concomitant increases in both cardiac contractile force and heart rate with simultaneous decreases in systolic and diastolic blood pressure. The intravenous infusion of milrinone caused parallel increases in both drug plasma concentration and cardiac contractile force; following termination of the milrinone infusion, there was a gradual decline in both its plasma concentration and in its inotropic activity, with a similar time course for these two parameters. A positive correlation (r = 0.78; p less than 0.008) was obtained between milrinone plasma concentration and its inotropic effect.

Characterization of the cardiotonic effects of milrinone, a new and potent cardiac bipyridine, on isolated tissues from several animal species.

The cardiotonic activity of milrinone (Win 47203), a potent analogue of amrinone, was demonstrated in isolated guinea pig, cat, rabbit, rat, and hamster atria and papillary muscles. Milrinone, in concentrations of 0.1-300 micrograms/ml, caused concentration-dependent increases in guinea pig papillary muscle and atrial developed tension with minimal increases in atrial rate. Compared with the in vitro inotropic activity of amrinone, milrinone was approximately 30 times more potent. The inotropic and chronotropic effects of milrinone do not appear to be mediated by the release of endogenous norepinephrine, by the direct stimulation of beta-adrenergic or histaminergic receptors, or through the stimulation of prostaglandin synthesis. The inotropic response of the guinea pig papillary muscles to isoproterenol was potentiated when isoproterenol was given at the peak effect of a minimally effective concentration of milrinone or after prolonged incubation with milrinone. No potentiation was observed when isoproterenol was administered at the peak effect of a high concentration of milrinone, which suggests that the positive inotropic action of milrinone may not be solely attributable to cyclic AMP phosphodiesterase inhibition.

Cardiotonic activity of milrinone, a new and potent cardiac bipyridine, on the normal and failing heart of experimental animals.

Milrinone (Win 47203) is a potent cardiac bipyridine with inotropic and vasodilator properties. Its effects were studied in anesthetized and unanesthetized dogs and in isolated cardiac tissues from guinea pigs. In the anesthetized dog, the intravenous injection of milrinone (0.01-0.1 mg/kg) increased cardiac contractile force (CF) (23 +/- 6.1 to 87 +/- 8.9%), maximum left ventricular pressure development (24 +/- 5.8 to 119 +/- 16.1%), and cardiac output (16 +/- 4.5 to 33 +/- 8.9%), with less than a 30% increase in heart rate (HR). Significant decreases in systolic and diastolic blood pressures were seen at 0.3-3 mg/kg i.v. Oral doses of milrinone (0.1-1.0 mg/kg), in unanesthetized dogs, increased cardiac CF by 35 +/- 7.0 to 99 +/- 17.0%, with a maximum increase in HR of 40 +/- 7.1% and no significant change in blood pressure. Milrinone was effective in the presence of ouabain and dopamine without enhancing their arrhythmogenic properties. It was also effective in reversing propranolol-, verapamil-, or pentobarbital-induced heart failure. The inotropic response does not seem to involve the stimulation of the autonomic receptors, the release of endogenous catecholamines, histamine, or prostaglandins, or the inhibition of Na+,K+-adenosine triphosphatase. Milrinone is an inhibitor or cardiac adenosine 3',5'-monophosphate (cAMP) phosphodiesterase, with resultant increases in cardiac cAMP levels. However, the time course for this increase does not seem to correspond to the increase in muscle developed tension, and, therefore, it is unlikely to be responsible for the initiation of the inotropic response. Milrinone is a potent cardioactive agent which should be beneficial in the treatment of acute and chronic congestive heart failure.

Relationship between amrinone plasma concentration and cardiac index.

Amrinone was given to 14 patients with congestive heart failure as an intravenous bolus (1 mg/sec) at doses ranging from 0.5 to 3.5 mg/kg. Simultaneous determinations of cardiac index were made by thermodilution and of amrinone plasma concentration by high-performance liquid chromatography. A relationship between improvement in cardiac index and increasing plasma concentrations of amrinone was demonstrated for 13 of the 14 patients. The percentage increase in cardiac index correlated with amrinone plasma concentration (r = 0.81; p less than 0.001). Amrinone was given to four patients as an intravenous bolus dose of 1.5 mg/kg followed by a constant infusion of 10 micrograms/kg/min for 10 hr; simultaneous determinations of cardiac index and circulating levels of amrinone indicated that both declined after the initial rise. The plasma concentration of amrinone remained relatively constant during the infusion at about 1.7 micrograms/ml. In all cases, despite the relatively constant plasma levels there was a decline in cardiac index after about 4 to 5 hr of infusion, although the cardiac index remained above the baseline; during the constant infusion the cardiac index rose again and was maintained at a reasonably constant level for the last 3 hr. Seven patients received oral doses of amrinone of about 3 mg/kg, and simultaneous determinations of cardiac index and plasma concentration showed a relationship between amrinone level and rise in cardiac index (p less than 0.05). In 16 patients after amrinone orally sufficient blood samples were taken to estimate the apparent first-order terminal elimination t 1/2. The t 1/2 as estimated by log-linear regression ranged from about 3 to 15 hr; mean +/- SEM value was 8.3 (+/- 1.1) hr.

Amrinone metabolism.

High-performance liquid chromatographic methods for the analysis of amrinone in plasma and for both amrinone and its N-acetyl metabolite in urine were developed and applied to measure specimens obtained from a number of healthy men who had received intravenous or oral amrinone. The intravenous doses ranged from 0.8 to 2.2 mg/kg. Terminal elimination of amrinone from the bloodstream followed apparent first-order kinetics. Half-life, after the drug had distributed to the tissues, was estimated by a log-linear least-squares regression; mean half-life was 2.6 +/- 1.4 hr. During the first 24 hr after medication, unchanged amrinone excreted in the urine of these subjects represented 10% to 40% of the dose. N-Acetyl metabolite in the urine represented less than 2% of the dose. In the oral study, doses ranged from 25 to 250 mg (0.31 to 3.5 mg/kg) and the maximum plasma concentration attained was proportional to the dose. The first order terminal elimination half-life was possibly dose-related. In only one subject were there unequivocal amounts of the N-acetyl metabolite in the plasma.

Cardiotonic activity of amrinone--Win 40680 [5-amino-3,4'-bipyridine-6(1H)-one].

The cardiotonic activity of a new, noncatechol, nonglycoside agent, amrinone, was investigated in vitro and in anesthestized and unanesthetized dogs. Amrinone (3-100 microgram/ml) caused a dose-dependent increase in papillary muscle developed tension and df/dt without significant changes in duration of the contractile cycle or time-to-peak tension. Amrinone induced slight increases in right atrial rate with no changes in electrophysiological properties of the cat papillary muscle or dog Purkinje fibers. In anesthetized dogs, intravenous bolus injections of amrinone at doses ranging from 1 to 10 mg/kg caused increases in cardiac contractile force and left ventricular dp/dt max with relatively small changes in heart rate and blood pressure. No significant changes in lead II ECG were observed. In unanesthetized dogs, intravenous infusion of amrinone (10-100 microgram/kg per min) caused increases in left ventricular dp/dt max and only small changes in heart rate and blood pressure. Amrinone, tested orally in this model at doses of 2-10 mg/kg, produced a positive inotropic effect with a rapid onset and long duration of action. The inotropic response to amrinone was not blocked by propranolol, dibenzyline, chlorisondamine, atropine, metiamide, or reserpine. Amrinone's inotropic response was not associated with significant alterations in cardiac norepinephrine, phosphodiesterase, cyclic AMP, or Na+, K+-activated ATPase.

Amrinone: a new non-glycosidic, non-adrenergic cardiotonic agent effective in the treatment of intractable myocardial failure in man.

Chronic congestive heart failure not controlled by conventional therapy was treated with intravenous amrinone, a new non-glycosidic, non-catecholamine cardiotonic agent. Eight patients with New York Heart Association functional class III-IV symptoms were hemodynamically monitored. At peak effect, cardiac index (CI) increased from 1.84 +/- 0.32 to 2.74 +/- 0.44 l/min/m2 (mean +/- SD) (p less than 0.001) and left ventricular filling pressure (LVFP) decreased from 25.8 +/- 6.2 to 19.5 +/- 6.8 mm Hg (p less than 0.05), while heart rate and mean aortic blood pressure did not change significantly. Mean endocardial circumferential fiber shortening (mean Vcf), determined by echocardiography, increased from 0.61 +/- 0.27 to 0.89 +/- 0.34 cir/sec (p less than 0.05). The duration of action after bolus infusion varied from 60--90 minutes. During continuous infusion of amrinone, sustained increases in CI and reductions in LVFP, similar to those at the time of peak effect after bolus administration, were maintained for 180 minutes. These marked cardiotonic effects of amrinone in patients already taking digitalis for severe heart failure occurred without side effects of arrhythmias or altered arterial pressures. The fact that the drug is orally active makes amrinone a v:ry promising inotropic agent for the treatment of chronic heart failure in man.

Hemodynamic assessment of amrinone. A new inotropic agent.

Amrinone, a new bipyridine derivative, exerts a positive inotropic action in experimental preparations and is effective when administered orally to dogs. To assess its immediate effects in man, we studied by cardiac catheterization the hemodynamic responses to amrinone (1.85 to 3.5 mg per kilogram given intravenously) in eight patients with congestive heart failure already receiving full doses of digitalis. the following statistically significant (P less than 0.01) effects were noted: cardiac index increased from a mean +/- 1 S.D. of 1.8 +/- 0.3 to 2.6 +/- 0.3 liters per minute per square meter; peak rate of left ventricular pressure rise rose from 849 +/- 233 to 1206 +/- 456 mm Hg per second; left ventricular end-diastolic pressure fell from 25 +/- 9 to 14 +/- 7 mm Hg; pulmonary-capillary pressure fell from 28 +/- 8 to 15 +/- 4 mm Hg; and right atrial pressure fell from 12 +/- 6 to 7 +/- 5 mm Hg. Mean heart rate was unchanged, and aortic mean pressure declined slightly (86 +/- 10 to 80 +/- 7 mm Hg, P less than 0.025). No toxicity was observed. Amrinone, whose mechanism of action has not yet ben defined, warrants further study as a possible treatment for heart failure.