Effects of digoxin on isolated human peripheral arteries and veins

Ouabain-induced coronary vasoconstriction in cats is not neurally mediated

1. Previous reports indirectly implicate a neural mechanism for coronary constriction to centrally administered digitalis. However, autoregulatory changes in coronary resistance due to changes in arterial pressure may have influenced the interpretation of these studies. 2. We tested directly the hypothesis that cardiac sympathetic innervation is responsible for coronary constriction to ouabain by examining the effects of ouabain (intravenous (i.v.) and intracerebroventricular (i.c.v.)) before and after bilateral stellate ganglionectomy. 3. Cats were anaesthetized and instrumented for the measurement of heart rate, blood pressure and coronary blood flow velocity using an epicardial-attached suction Doppler probe. Animals were treated with atenolol and the effects of either i.v. or i.c.v. injections of ouabain were examined. 4. In seven cats treated with atenolol, i.v. ouabain (0.11 mg/kg) produced maximal increases in arterial pressure and coronary vascular resistance index (CVRI) of 66 +/- 7 mmHg and 37 +/- 9%, respectively. Following bilateral stellate ganglionectomy (n = 7), ouabain produced similar increases in arterial pressure (70 +/- 9 mmHg) and CVRI (39 +/- 7%). A higher dose of i.v. ouabain (1.1 mg/kg) produced maximal increases in arterial pressure (115 +/- 4 mmHg) and coronary resistance (86 +/- 14%) in intact cats (n = 6) that were similar to responses seen in cats in which stellate ganglionectomy had been performed (n = 6; arterial pressure 104 +/- 13 mmHg; coronary resistance 114 +/- 6%). The increases in coronary resistance to ouabain at both doses were significantly greater than increases in coronary resistance to passive elevation of arterial pressure during aortic constriction. Thus, autoregulation does not explain fully the coronary constriction to ouabain. 5. To further examine a central mechanism, i.c.v. perfusion with 0.3 mmol/L ouabain was performed in six cats, resulting in increases in arterial pressure (122 +/- 7 mmHg) and coronary resistance (58 +/- 14%). Similar increases in arterial pressure (117 +/- 16%) and coronary resistance (84 +/- 20%) were seen in separate studies (n = 6) following stellate ganglionectomy. 6. These results indicate that coronary constriction to ouabain does not require intact cardiac sympathetic innervation, but probably involves a direct or humorally mediated effect.

Effects of digoxin, furosemide, enalaprilat and metoprolol on endothelial function in young normotensive subjects

1. Endothelial dysfunction is seen in patients with essential hypertension or congestive heart failure (CHF). The present study aimed to evaluate the direct effect on endothelium- dependent vasodilation (EDV) of different pharmacological drugs commonly used in the treatment of these conditions. 2. Forearm blood flow (FBF) was measured in 37 young healthy normotensive subjects with venous occlusion plethysmography during local intra-arterial infusions of methacholine (MCh; 2-4 microg/min), evaluating EDV, and sodium nitroprusside (SNP; 5-10 microg/min), evaluating endothelium-independent vasodilation (EIDV). The measurements of EDV and EIDV were undertaken under baseline conditions and were repeated after 1 h intra-arterial infusion of digoxin (0.1 mg/h), furosemide (5.0 mg/h), enalaprilat (2,4 mg/h), metoprolol (1.2 mg/h) or saline (controls). 3. Enalaprilat and digoxin improved the FBF response to MCh at 4 microg/min (from 22.7+/-2.3 to 25.5+/-2.1 mL/min per 100 mL tissue (P < 0.01) and from 18.2+/-2.4 to 22.2+/-2.0 mL/min per 100 mL tissue (P < 0.05), respectively). No significant changes where induced by furosemide or metoprolol in response to MCh at 4 microg/min (from 19.4+/-2.0 to 22.9+/-2.8 and from 15.3+/-2.4 to 14.7+/-1.1 mL/min per 100 mL tissue, respectively). No significant changes in basal FBF or EIDV were induced by the different drugs. When the endothelial function index was calculated as the MCh: SNP FBF ratio, a significant improvement was seen only with enalaprilat (1.1+/-0.1 to 1.2+/-0.1; P < 0.01) and furosemide (1.0+/-0.1 to 1.3+/-0.4; P < 0.05). 4. In conlusion, the results of the present study show that enalaprilat and furosemide improve endothelial vasodilatory function, while no major effect was induced by digoxin or metoprolol. Thus, different direct effects on the endothelium in young normotensive subjects were induced by drugs commonly used in the treatment of hypertension or CHF.

Protection from digoxin-induced coronary vasoconstriction in patients with coronary artery disease by calcium antagonists

Intravenous digoxin induces constriction of normal and stenotic coronary arteries in patients with coronary artery disease, which may lead to ischemic complications. We found that pretreatment with oral nisoldipine and intracoronary nitroglycerin neutralizes this digoxin-induced effect.

Effects of ouabain on the isolated human uteroplacental vasculature

OBJECTIVE

We sought to describe the effects of ouabain on the human uteroplacental vasculature.

STUDY DESIGN

Stem villous vessels and intramyometrial arteries isolated from placental and myometrial biopsy specimens at term were mounted in organ baths. Moreover, isolated human placental cotyledons were perfused.

RESULTS

Contractions induced by the thromboxane A2 analog U46619 were unaffected by pretreatment with ouabain 10(-10) to 10(-6) mol/L. In fetal vessels nitric oxide (10(-8) to 3 x 10(-5) mol/L) induced relaxation of vascular tonus induced by U46619 and potassium. This relaxation was inhibited by pretreatment with ouabain 10(-7) to 10(-6) mol/L. These associations were unaffected by removal of the endothelium. In maternal arteries ouabain (10(-6) mol/L) failed to significantly affect nitric oxide-induced relaxation. Ouabain (10(-9) mol/L) significantly affected pressure-flow relationships in perfused cotyledons.

CONCLUSIONS

Ouabain impairs nitric oxide-induced relaxation of human stem villous arteries and veins, which may explain the changes induced by therapeutically relevant concentrations of the drug on pressure-flow relationships in the perfused cotyledon. Thus treatment with cardiac glycosides in pregnancy may impair uteroplacental blood flow.

Nonocclusive mesenteric ischemia

NOMI remains a challenging clinical entity that demands a heightened index of suspicion and an aggressive diagnostic and treatment strategy in order to avoid the untoward sequela of short bowel syndrome and to achieve survival. Early arteriographic diagnosis and prompt institution of vasodilator therapy have proven successful in reducing the high (70% to 90%) mortality rate observed through the 1980s to 50% to 55% during the last decade. Continued investigations into the pathophysiologic mechanisms underlying splanchnic vasospasm and intestinal IR injury will, it is hoped, add more alternate and effective therapies to the current armamentarium.

Effects of two endogenous Na+,K(+)-ATPase inhibitors, marinobufagenin and ouabain, on isolated rat aorta

Previously, we reported that the venom of Bufo marinus toad contains a Na+,K(+)-ATPase inhibitor with potent vasoconstrictor activity. In the present study, using thin-layer chromatography in Silicagel 60 F254 + 366, we separated a vasoactive substance from a mixture of steroids from Bufo marinus venom. Based on chromatographic mobility of this substance and typical color reaction after its vizualization with SbCl3, we identified it as a previously described steroid, marinobufagenin. Vasoconstrictor and Na+,K+ pump inhibitory properties of marinobufagenin were studied in isolated rat aortic rings and compared with those of ouabain. Ouabain (10-100 mumol.1-1) produced weak vasoconstriction, which was blocked by 2 mumol.1-1 phentolamine. 10 mumol.1-1 ouabain stimulated, and at higher concentrations inhibited, the Na+,K+ pump. 2 mumol.1-1 phentolamine abolished the activating effect of 10 mumol.1-1 ouabain on the Na+,K+ pump, but did not alter the inhibitory action of higher concentrations of ouabain. By contrast, marunibufagenin elicited rapid and strong vasoconstriction and inhibited ouabain-sensitive 86Rb uptake. Antidigoxin antibody antagonized the vasoconstrictor responses to marinobufagenin, but not to ouabain. 2 mumol.1-1 phentolamine did not alter the constrictor effect of marinobufagenin. In solid-phase digoxin immunoassay, marinobufagenin demonstrated higher digoxin-like immunoreactivity than ouabain.

Mechanical and electrophysiological studies on the positive inotropic effect of 2-phenyl-4-oxo-hydroquinoline in rat cardiac tissues

1. The pharmacological and electrophysiological effect of 2-phenyl-4-oxo-hydroquinoline (YT-1), a new synthetic agent, were determined in rat isolated cardiac tissues and ventricular myocytes. 2. YT-1 was found to have a positive inotropic effect in both atria and ventricular muscles but did not cause significant increases in the spontaneously beating rate of right atria. 3. The positive inotropic effect of YT-1 was antagonized neither by beta-nor by alpha-adrenoceptor antagonists but was partially antagonized by a Ca2+ channel blocker (verapamil) and a K+ channel blocker (4-AP). 4. The action potential duration and amplitude of ventricular cells were progressively increased as the concentration of YT-1 was increased from 3 to 30 microM. 5. A voltage clamp study revealed that the prolongation of action potential duration by YT-1 was associated with a prominent inhibition of 4-AP-sensitive transient outward current (I(to)). At potentials negative to the reversal potential of K1-channels, the inward current through these channels was partially reduced by YT-1. At potentials positive to the reversal potential, the outward current through these channels was affected very little. 6. Although YT-1 blocked the amplitude of I(to), the voltage-dependence of the steady-state inactivation of I(to), was unaffected. 7. Apart from the inhibition of K+ currents, YT-1 also inhibited the sodium inward current. 8. The evidence suggests that YT-1 increases the slow inward Ca2+ current (ICa) significantly. 9. It is concluded that the positive inotropic effect of YT-1 is due predominantly to the increase of ICa and inhibition of I(to).

Endothelial role in ouabain-induced contractions in guinea pig carotid arteries

The influence of endothelium on the direct contractile effects of ouabain in vascular smooth muscle was analyzed in isolated perfused guinea pig carotid arteries. After blocking the neurogenic component of the glycoside contraction with alpha-adrenergic receptor blocking drugs or treating the animals with reserpine, ouabain-induced contractions were markedly reduced in vessels with intact endothelium. However, removal of the vascular endothelium from reserpinized carotid arteries resulted in ouabain-induced contractions similar to those observed in control arteries. These effects were not mimicked by the inhibitor of nitric oxide NG-monomethyl L-arginine or by the cyclooxygenase blocker indomethacin. Bioassay experiments suggested that these endothelial effects are mediated by diffusible factors. Uptake of 86Rb to measure sodium pump activity was significantly reduced by removal of the endothelium. These results suggest the existence of an inhibitory modulation by the endothelium of contractions induced by ouabain, likely mediated by a diffusible factor (or factors) released from these cells. The nature of this substance is unknown, but it is neither related to prostaglandins nor a nitric oxide-related compound. Its mechanism of action could be the stimulation of vascular sodium pump activity, the antagonism of the pump's inhibition by ouabain, or both.

Endothelial modulation of the ouabain-induced contraction in human placental vessels

Ouabain (1 x 10(-7)-3 x 10(-5) M) elicited concentration-dependent vasoconstriction in human placental arteries and veins. These responses, but not those produced by 10(-6) M 5-hydroxytryptamine, were increased after the removal of vascular endothelium. In placental arteries, the respective blockade of cyclooxygenase or lipoxygenase with indomethacin of 5,8,11,15-eicosatetraynoic acid as well as the inactivation of nitric oxide with phenidone or oxyhemoglobin and the inhibition of nitric oxide synthesis with NG-monomethyl L-arginine (all at 10(-5) M) did not mimic the effects of endothelial denudation on ouabain-evoked contractions. Bioassay experiments suggested that the above-mentioned endothelial effects are mediated by diffusible factors. 86Rb+ uptake, a method to measure sodium pump activity, was significantly reduced by the removal of endothelium. These results suggest the existence of an inhibitory modulation by the endothelium of the contractions induced by ouabain, likely mediated by a diffusible factor(s) released from these cells. The nature of this substance is unknown but is not related to prostaglandins or leukotrienes, and neither is it a nitric oxide-related compound. Its mechanism of action could be stimulating the activity of vascular sodium pump and/or antagonizing its inhibition by ouabain.

Control of arteriolar resistance in heart failure. Partial attenuation of specific phosphodiesterase inhibitor-mediated vasodilation by digitalis glycosides

BACKGROUND

The vasodilatory response to local specific type III phosphodiesterase inhibition with amrinone was evaluated before and immediately after local administration of digoxin in 14 patients with severe congestive heart failure (CHF).

METHODS AND RESULTS

A 3F polyethylene catheter was inserted into the common femoral artery for drug administration and pressure monitoring. Mean blood flow velocity (MBFV) was continuously determined in the superficial femoral artery by transcutaneous Doppler ultrasonography. After intra-arterial administration of 10 mg amrinone, group MBFV increased from 7.7 +/- 1.4 to 16.0 +/- 2.1 cm/sec (p less than 0.05, n = 10). Local administration of 20 micrograms digoxin, which was infused over 20 minutes, did not alter group MBFV (i.e., 8.2 +/- 1.6 versus 7.6 +/- 1.5 cm/sec; p = NS, n = 10). The second administration of 10 mg amrinone, which immediately followed completion of local digoxin infusion, increased group MBFV but to a lesser extent than that produced by the first amrinone administration (i.e., 11.9 +/- 1.9 versus 16.0 +/- 2.1 cm/sec; p less than 0.05, n = 10). When placebo was administered instead of digoxin, group MBFV was similar after the first and second administrations of amrinone (i.e., 15.3 +/- 3.3 versus 15.6 +/- 3.8 cm/sec; p = NS, n = 4).

CONCLUSIONS

Although local administration of digoxin did not significantly alter baseline vascular tone in patients with CHF, it substantially decreased the direct vasodilatory effect induced by specific type III phosphodiesterase with amrinone.

Effects of ouabain and low sodium on contractility of human resistance arteries

Earlier work with rat arteries has resulted in a widely held assumption that resistance artery smooth muscle will not contract on exposure to a reduced transplasmalemmal sodium gradient. In view of the well-recognized low sensitivity of rat tissue to cardiac glycosides, we have investigated the effects of altering the transplasmalemmal sodium gradient on vascular smooth muscle tone by using human resistance arteries. Incubation of arteries in low sodium or in ouabain to inhibit active sodium efflux for 1 hour increased the contractile response to caffeine stimulation; this finding indicated enhanced calcium buffering by the sarcoplasmic reticulum. Prolonged incubation in ouabain in the presence of phentolamine or diltiazem resulted in a concentration-dependent increase in the tone of resting human resistance arteries. Reduction of the transplasmalemmal sodium gradient by incubation in low sodium buffer effected an increase in tone similar to that obtained in the presence of ouabain. These results suggest that alteration of the transplasmalemmal sodium gradient may increase the vascular smooth muscle tone of human resistance arteries by altering intracellular calcium handling. This is a new finding in human resistance arteries and may involve inhibition and, indeed, reversal of sodium-dependent calcium efflux. A concentration-dependent potentiation of tone was found after the addition of ouabain to submaximally activated arteries. Sodium-calcium exchange may also play a pivotal role in this mechanism.

Relationship between systemic and coronary vascular responses to digoxin and concurrent drug therapy with verapamil/beta-adrenoceptor antagonists in humans

1. In 24 patients who were undergoing coronary arteriography for the assessment of ischaemic heart disease, the relationship between the systemic and coronary vascular responses to acute intravenous digoxin administration (500 micrograms) and concurrent drug therapy with the calcium antagonist verapamil (group I) or a beta-adrenoceptor antagonist (group II) or neither of these agents (group III) was examined. 2. Systemic vascular resistance (SVR) tended to rise rapidly after digoxin injection in patients in groups II and III, and tended to decline initially in patients in group I; however, these differences were not statistically significant (variance ratio [VR] = 0.77). 3. No significant differences were observed in coronary vascular responses to acute digoxin administration between the three groups of patients (VR = 0.34). 4. For the entire group of 24 patients, no statistically significant digoxin-induced effects on resistance could be demonstrated in either the systemic or coronary circulations, although in individual patients vasoconstrictor effects were observed. 5. We conclude that acute intravenous administration of digoxin does not consistently cause systemic or coronary vasoconstriction in patients with ischaemic heart disease. Variability in vasomotor responses to digoxin is not clearly related to concurrent drug therapy with verapamil or a beta-adrenoceptor antagonist. The observation that systemic vascular resistance tends to increase in the first few minutes after digoxin injection should be addressed in future studies.

Effects of ouabain on isolated cerebral and femoral arteries of the cat: a functional and biochemical study

1. This study analyzes the mechanisms involved in the responses to ouabain in cat cerebral and femoral arteries and characterizes the electrogenic Na+ pump present in these vessels. The latter was accomplished by measurement of [3H]-ouabain binding to arterial membrane fractions, K+-elicited relaxation and ouabain-sensitive 86Rb+ uptake. 2. Ouabain induced transient contraction in cylindrical segments of cerebral arteries. This contraction was reduced by verapamil (3 X 10(-6) M) and Ca2+-removal from the medium but was not modified by phentolamine (3 X 10(-6) M) or pretreatment with reserpine. However, the contraction elicited by ouabain in femoral artery segments lasted longer, and was reduced by Ca2+-omission, phentolamine or reserpine, but remained unaffected by verapamil. 3. The immersion of the arteries in low-Na+ (25 mM) medium abolished the contraction caused by ouabain. 4. The exposure of the arteries to a K+-free medium induced a small transient increase in tension, and the subsequent application of K+ (7.5 mM) elicited a marked relaxation. This effect was greater in cerebral than in peripheral arteries, and was suppressed by ouabain (10(-4) M). 5. Scatchard analysis of the [3H]-ouabain binding to arterial membrane fractions suggested a single class of binding sites. The KD values for both kinds of arteries were of similar order, while the Bmax value was greater in cerebral than in femoral arteries. 6. Total and ouabain-sensitive 86Rb+ uptakes were greater in cerebral than in femoral vessels. 7. These results indicate that: (1) ouabain-induced contraction of cerebral arteries is due to a direct effect on vascular smooth muscle cells, while in femoral arteries it is due to noradrenaline release from adrenergic nerve terminals; and (2) the electrogenic Na+ pump activity is greater in cerebral than in peripheral arteries.

Digoxin inhibition of relaxation induced by prostacyclin and vasoactive intestinal polypeptide in small human placental arteries

Small chorionic plate arteries were obtained from human placentae following normal vaginal delivery. Tubal vascular preparations were dissected, mounted in organ baths, and their isometric tension was recorded. Digoxin (10(-6) M) caused a rise in basic tension, reaching a maximum of 17 per cent of contractions induced by potassium (124 mM) depolarization. Pretreatment with digoxin did not significantly influence the concentration-dependent contractile responses to 5-hydroxytryptamine and prostaglandin F2 alpha (PGF2 alpha). In preparations contracted with PGF2 alpha, cumulative addition of prostacyclin (PGI2) and vasoactive intestinal polypeptide (VIP) produced concentration dependent relaxations. Digoxin (10(-8) to 10(-6) M) inhibited and finally abolished these relaxant effects of PGI2 and VIP in a concentration-dependent fashion. Pretreatment by digoxin (10(-8) to 10(-6) M) diminished the relaxant effect of sodium nitroprusside, but the effect was less pronounced than that on PGI2- and VIP-induced relaxation. As PGI2 and VIP may be of importance for the maintenance of a low resistance of the fetal placental vascular bed, the finding that digoxin decreases the vasodilating effects of these agents might imply effects on placental resistance of cardiac glycosides when used in late human pregnancy.

Extracardiac and coronary vascular effects of digitalis

The administration of digitalis glycosides causes a variety of extracardiac effects. In both normal human subjects and in other species, digitalis increases smooth muscle tone of resistance and capacitance vessels. The vasoconstriction is mediated, in part, by a direct action of these glycosides on smooth muscle and, in part, by an increase in alpha-adrenergic tone. Constriction of coronary and splanchnic vessels may lead to myocardial or mesenteric ischemia. In contrast to normal subjects, patients with congestive heart failure demonstrate arteriolar and venodilation in response to these glycosides, possibly because the myocardial effect, to increase cardiac output and peripheral blood flow, overcomes the vasoconstrictor properties of these drugs. Other important actions of digitalis glycosides occur in the central and peripheral nervous systems. Their effects on the area postrema of the medulla oblongata are largely responsible for the alpha-adrenergic-mediated peripheral vasoconstriction, as well as the nausea and vomiting that frequently accompany digitalis intoxication. Actions of glycosides on the cerebral cortex are responsible for the wide range of neurotoxic effects that range from visual disturbances and headaches to seizures and coma. Finally, peripheral neurologic effects of digitalis glycosides on baroreceptor and cardiac afferent fibers may: improve the depressed function of these receptors in the situation of heart failure, and reflexly lower peripheral vascular resistance, thereby partially preventing the vascular constrictor action of these glycosides.

Influence of atenolol and nifedipine on digoxin-induced inotropism in humans

Short term effect of digoxin on left ventricular performance was studied in six healthy volunteers before and during atenolol or nifedipine administration. Left ventricular function was evaluated by systolic time intervals and echocardiography. No changes in left ventricular end diastolic or systolic dimensions occurred throughout the study, indicating unchanged ventricular pre- and afterload. Thus, changes in the systolic time intervals must be attributed to changes in cardiac contractility. Changes in the pre-ejection period index (PEPI) obtained from the systolic time intervals were used as a measure of digoxin-induced inotropism. A concentration-response relationship between plasma digoxin level and changes in PEPI was revealed when digoxin was given alone. Atenolol did not influence the digoxin-induced inotropism at a given serum digoxin level. During nifedipine administration no inotropic effect of digoxin could be demonstrated. Thus, it is concluded that nifedipine attenuates digoxin-induced inotropism, while atenolol seems without this effect. These results are in accordance with previous experiments and reflect the different pharmacological sites of action of beta-adrenoceptor antagonists and calcium channel blocking agents. Plasma digoxin concentration, renal digoxin clearance and creatinine clearance did not change during atenolol or nifedipine.

Potentiating and depressive effects of ouabain and potassium-free solutions on rat mesenteric resistance vessels

We have investigated the in vitro effects of ouabain and K-free solutions on some pharmacological and electrophysiological properties of rat mesenteric resistance vessels (internal diameter approximately 190 micrometers). Vessels were mounted as ring preparations on a myograph capable of measuring their isometric wall tension. In normal saline solutions, vessels did not exhibit any tone and had a membrane potential of -54 mV. Both 1 mM ouabain and K-free solutions caused a transient depolarization of 5-8 mV; thereafter the membrane slowly depolarized to about -45 mV after 30 minutes. There was no mechanical response to ouabain, but K-free solutions caused a transient development of tension which could be inhibited by phentolamine (1 microM). In norepinephrine-activated vessels, exposure to ouabain or K-free solutions caused a small depolarization and an increase in tension. Long-term (30-minute) exposure to 1 mM ouabain or K-free solutions reduced the amplitude of norepinephrine responses and, for the lower (but not the higher) norepinephrine concentrations, the membranes were about 14 mV more depolarized than control. The mechanical responses to a cocktail of norepinephrine in a high potassium solution were, however, unaffected. Re-exposure to normal saline solution produced a transient hyperpolarization and transiently eliminated the norepinephrine response, but thereafter the membrane potential and response returned to normal. The results indicate that ouabain and K-free solutions can have both short-term potentiating and long-term depressive effects on the mechanical response of rat mesenteric resistance vessels to norepinephrine.