Physiology of the Failing Heart

The treatment of both acute and chronic heart failure is based on our current understanding of the multitude of changes that occur in this condition. The major underlying cause of heart failure is an impairment of myocardial contractile function. The associated decrease in stroke volume and cardiac output initiates a multifaceted sequence of neurohormonal and vascular events that affect preload, afterload, and heart rate. For many years heart failure was attributed to left ventricular dysfunction, which was corrected by use of positive inotropic agents, such as digoxin. It is now clear that heart failure represents a highly complex series of events that include neuroendocrine activation. This has resulted in reevaluation of the approaches to management. Whereas the hemodynamic model may still apply to patients with acute failure, the new model focuses on prevention of progression in the outpatient setting by preventing or delaying the development of left ventricular remodeling.

Left Ventricular Remodeling

Remodeling of the heart occurs through complex structural changes in one or more cardiac chambers, especially the ventricles. These result in an increase in end-diastolic and end-systolic volume along with changes in cardiac shape and left ventricular mass (Frank-Starling curves are shown in Figure 23-1). Impaired contractility was previously thought to be responsible for heart failure, although a specific biochemical abnormality could not be identified. This idea has given way to the concept that heart failure involves endogenous neurohormones and cytokines in response to an initial “index event,” usually an acute injury to the heart or genetic mutation. Coronary artery disease and hypertension account for most cases, with myocardial infarction being a major contributor. Any insult, whether acute myocardial infarction, essential hypertension, aortic stenosis, or volume overload caused by aortic insufficiency, idiopathic cardiomyopathy, or inflammatory disease, leads to activation of specific mediators involved in the remodeling process.

FIGURE 23–1 Frank-Starling ventricular function curve. Force of contraction, expressed as left ventricular dP/dt (rate of pressure development during early systolic phase) is a function of (1) left ventricular volume before the onset of contraction or (2) sarcomere length. The length of the sarcomere, the unit between two Z lines of a myofibril, in the normal heart is 1.7 to 1.8 μm at the endocardial and epicardial layers and 2.0 μm in the middle layer. The reduced force of contraction in a failing heart is partly compensated for by an increase in the end-diastolic volume, which increases the length of the sarcomere, thereby increasing the force of contraction or the stroke volume (A, arrow). Positive inotropic interventions in the normal heart are canceled by shortening of the muscle (B, arrow). Digitalis glycosides shift the ventricular function curve and reduce the end-diastolic volume required for the muscle to develop the necessary force of contraction (C, arrow).

Neurohormonal systems defend against changes in intravascular volume and act to maintain regional blood flow and regulate systemic blood pressure. Initially, they compensate for the decline in ventricular function and mask the underlying deficiency. Chronic activation of multiple compensatory mechanisms perpetuates progression to irreversible myocyte injury and worsening of cardiac function. Initiating factors include: stretch of the ventricular myocardium, increased cytokine and growth factor production, nitric oxide (NO) production, tumor necrosis factor, natriuretic peptides, free radical-mediated oxidative stress, ischemia, activation of myocardial metalloproteinases, proapoptotic factors, and chronic inflammation. Heart failure is usually accompanied by an increase in sympathetic nervous system (SNS) activation along with chronic up regulation of the renin-angiotensin-aldosterone system (RAAS) and effects of aldosterone on heart, vessels, and kidneys. CHF should be viewed as a complex, interrelated sequence of events involving hemodynamic, nonhemodynamic, genetic, energetic, and neurohormonal events.

Sympathetic Nervous System

In the failing heart, the loss of contractile function leads to a decline in cardiac output and a decrease in arterial blood pressure. The baroreceptors sense the hemodynamic changes and initiate countermeasures to maintain support of the circulatory system. Activation of the SNS serves as a compensatory mechanism in response to a decline in left ventricular stroke volume. This helps maintain adequate cardiac output by increasing myocardial contractility and heart rate (β₁ adrenergic receptors) and by increasing vasomotor tone (α₁ adrenergic receptors) to maintain systemic blood pressure (see Chapters 11 and Chapters 19). Over the long term, this hyperadrenergic state leads to irreversible myocyte damage, cell death, and fibrosis. In addition, the augmentation in peripheral vasomotor tone increases left ventricular afterload, placing an added stress upon the left ventricle and an increase in myocardial O₂ demand, factors involved in ventricular remodeling. The frequency and severity of cardiac arrhythmias are enhanced in the failing heart, in part as a result of the increased adrenergic tone.

Systolic Heart Failure

The most frequent cause for chronic systolic dysfunction is ischemic cardiomyopathy, characterized by a reduction in the ventricular ejection fraction and enlargement of the left ventricle, due to a failure of the left ventricle to empty as a result of impaired myocardial contractility or pressure overload. This results from destruction of myocytes, impaired myocyte function, or fibrosis. Chronic pressure overload, caused by untreated long-standing hypertension or aortic stenosis, decreases the left ventricular ejection fraction by increasing resistance to forward flow. Initially, the increase in the left ventricular end-diastolic pressure (volume) results in a compensatory enhancement in stroke volume due to the pressure-induced lengthening of the sarcomeres that invokes the Frank-Starling mechanism (see Fig. 23-1), thereby partially compensating for the failing ventricle. The marked diastolic derangement in filling and the decreased ventricular distensibility do not permit adequate stretch of the myocytes because it occurs under conditions that require an increase in cardiac output. Therefore, in the presence of systolic heart failure, the Frank-Starling mechanism fails to adequately increase stroke volume in response to exercise.

The compensatory increase in sympathetic tone plus the activation of the RAAS system maintains arterial blood pressure. However, homeostatic mechanisms also increase total peripheral resistance (left ventricular afterload). Circulating blood volume is also increased, further contributing to maintenance of arterial pressure. Over the long term, this places an additional burden on the failing heart. Moreover, because increases in peripheral resistance decrease tissue perfusion for a given blood pressure, and tissue perfusion is more important than blood pressure, these mechanisms may not be advantageous in the long term.

Energy efficiency is reduced in chronic heart failure because a greater wall tension is required to develop the necessary intraventricular pressure, and peripheral resistance is increased. Energetic efficiency decreases further when relaxation is inhibited in the hypertrophied heart and when the heart rate is increased by activation of the SNS, resulting in a reduced stroke volume.

Chronic pressure overload, as in uncontrolled hypertension, contributes to hypertrophy. Stretching of the sarcolemma, the ensuing influx of Na⁺, and increased angiotensin II concentrations in plasma are possible causes. The number of myocardial cells does not increase in adults, but each cell enlarges. Remodeling may occur in association with such hypertrophy. Remodeling involves a shift of isoforms of functional proteins, such as myosin, creatine kinase, and Na⁺,K⁺-adenosine triphosphatase (ATPase). These are adaptive events but may ultimately contribute to ventricular remodeling. Furthermore the hypertrophied heart loses compliance (i.e., the ability to relax).

Diastolic Heart Failure

Diastolic dysfunction may result from impaired early diastolic relaxation, increased stiffness of the ventricular wall, or both. Hypertension, valvular disease, or congenital abnormalities lead to development of diastolic dysfunction, with hypertrophied or poorly compliant ventricular walls impeding filling of the left ventricle. Diastolic relaxation is an energy-dependent process, which is impaired by myocardial ischemia and a temporary loss of energy production. In this case ventricular filling can be achieved only at a greater than normal filling pressure because of reduced left ventricular wall compliance. Abnormalities may be attributable in part to an increase in interstitial connective tissue. Hypertrophic myocytes exhibit abnormal Ca⁺⁺ cycles characterized by prolonged Ca⁺⁺ transients and impaired relaxation. Current evidence implicates the local activity of angiotensin II and elevated circulating levels of aldosterone, both of which are implicated in the development of myocardial fibrosis. They lead to deposition of excessive amounts of collagen and a decrease in ventricular compliance, increased chamber stiffness, or a decrease in distensibility. Patients with diastolic dysfunction may have a normal cardiac output, suggesting that pharmacological management will differ from patients with systolic heart failure.

Angiotensin-Converting Enzyme Inhibitors

Baroreceptor-mediated activation of the SNS leads to an increase in renin release and formation of angiotensin II (see Chapter 19), which causes intense vasoconstriction and stimulates aldosterone production (Fig. 23-2). This is decreased by the angiotensin-converting enzyme (ACE) inhibitors, which inhibit formation of angiotensin II from angiotensin I, as discussed in Chapter 20.

FIGURE 23–2 Renal release of renin leading to the formation of angiotensin II and subsequent activation of AT₁ receptor-mediated events.

Angiotensin II acts through AT₁ and AT₂ receptors, although most of its actions occur through AT₁ receptor activation. Although the AT₂ receptor is distributed widely in fetal tissues, its distribution is limited in adults. Angiotensin II mediates cell growth, vasoconstriction, Na⁺ and fluid retention, and sympathetic activation (Table 23-1). The central role of the RAAS in the development and progression of cardiovascular disease and, in particular, CHF is well established. In addition to regulation of blood pressure and maintenance of fluid and electrolyte balance, short-term activation of the RAAS in heart failure improves cardiac output through fluid and Na⁺ retention (increased preload), whereas long-term activation results in vasoconstriction, increased afterload, and decreased cardiac output. These mechanisms, together with SNS activation, induce a vicious cycle of increased preload, afterload, and cardiac workload, leading to increased myocardial O₂ consumption, loss of myocytes through apoptosis, and progressive worsening heart failure.

TABLE 23–1 Effects of Angiotensin II

β Adrenergic Receptor Blocking Drugs

There is overwhelming evidence to support the use of β receptor blockers in CHF; however, the mechanisms involved remain unclear. Part of the beneficial effects of these agents may derive from slowing of heart rate, which would improve coronary blood flow and decrease myocardial O₂ consumption. This would lessen the frequency of ischemic events and potential for development of a lethal arrhythmia. Activation of the SNS can provoke arrhythmias by increasing cardiac automaticity, increasing triggered activity leading to ventricular arrhythmias, which may account for the ability of β receptor blockers to reduce the incidence of sudden cardiac death in patients with ischemic heart disease and heart failure.

The β receptor blockers inhibit the adverse effects of the SNS in patients with heart failure. Whereas cardiac adrenergic drive initially serves as a compensatory mechanism to support the failing heart, long-term activation leads to a down regulation of β₁ receptors and an uncoupling from adenylyl cyclase (see Chapter 1), thereby reducing myocardial contractility. The β receptor blockers may be beneficial through resensitization of the down regulated receptor, improving myocardial contractility.

Angiotensin Receptor Blockers

Another approach is to block AT₁ receptors with the use of angiotensin receptor blockers (ARBs). The currently available drugs selectively block AT₁ receptors and replicate many of the actions of ACE inhibitors; they do not block AT₂ receptors. Activation of AT₂ receptors may cause vasodilation, preventing hypertrophy of vascular smooth muscle and cardiomyocytes, production of bradykinin, and release of NO. There is also overexpression of AT₂ receptors in the failing heart. There may also be non-ACE-dependent formation of angiotensin II by enzymes, such as chymase, cathepsin G, trypsin, and tissue plasminogen activator. An ARB might block the deleterious actions mediated by AT₁ receptors while preserving the desirable effects of AT₂ receptor activation. Although the hemodynamic and clinical effects may appear similar, ACE inhibitors and ARBs should not be regarded as being identical.

Aldosterone Antagonists

The elevated circulating angiotensin II levels in the patient with CHF lead to greatly increased production of aldosterone (see Chapter 39), an important mediator in the progressive development of CHF. Aldosterone binds to mineralocorticoid receptors in renal epithelial cells and promotes Na⁺ retention, Mg⁺⁺ and K⁺ loss, sympathetic activation, parasympathetic inhibition, myocardial and vascular fibrosis, baroreceptor dysfunction, impaired arterial compliance, and vascular damage. Aldosterone antagonists include spironolactone and eplerenone, which may reduce norepinephrine (NE) release from cardiac sympathetic nerves and increase plasma K⁺. The elevated concentrations of aldosterone in CHF led to the concept that inhibition of aldosterone receptors could be beneficial, and competitive aldosterone antagonists are now part of the therapeutic armamentarium.

Cardiac Glycosides

The cardiac glycosides increase the force of myocardial contraction, alter electrophysiological properties in specialized regions, and have extracardiac actions associated with toxicity. Cardiac glycosides influence the heart through a direct inhibition of membrane Na⁺,K⁺-ATPase and an indirect increase in vagal tone (Table 23-2). Their cardiotoxic effects are an overextension of the same mechanisms responsible for their positive inotropic actions.

TABLE 23–2 Effects of Cardiac Glycosides on Electrophysiological Properties of the Heart

Cardiac glycosides increase contractile force; however, unlike catecholamines, they do not increase the rate of relaxation. By decreasing the activity of the Na⁺,K⁺-ATPase, they cause a progressive gain in intracellular Na⁺ with each cardiac cycle. This increase promotes Ca⁺⁺ influx by Na⁺/Ca⁺⁺ exchange (Fig. 23-3). The net result is an increased intracellular Ca⁺⁺, enhancing the Ca⁺⁺ transient resulting from an augmented Ca⁺⁺ loading of the sarcoplasmic reticulum. In the presence of a digitalis glycoside, a new steady state is achieved where an increased amount of Ca⁺⁺ is released after depolarization, increasing force development (stroke volume; see Fig. 23-1).

FIGURE 23–3 Membrane ion flux of Na⁺ and Ca⁺⁺ in heart. Glycoside-induced inhibition of Na⁺,K⁺-ATPase secondarily promotes Na⁺/Ca⁺⁺ exchange.

Electrophysiological effects of cardiac glycosides vary among different regions of the heart. They decrease automaticity within the sinoatrial (SA) and atrioventricular (AV) nodes as a result of an increase in parasympathetic tone along with a concomitant decrease in sympathetic tone. The increase in parasympathetic tone on the AV node leads to a decrease in conduction velocity and an increase in effective refractory period. Thus digitalis glycosides indirectly decrease heart rate and impair impulse transmission across the AV node.

The major direct effects are in the atrial muscle, AV node, and ventricles (see Table 23-2). In atria, they prolong the effective refractory period and decrease conduction velocity, effects opposite to those elicited by their indirect actions. However, the direct effects in the AV node summate with the indirect actions to further impair conduction velocity and increase the refractory period.

There are also significant extracardiac effects of digitalis glycosides. Most cells express membrane Na⁺,K⁺-ATPases, but those in excitable tissues have a higher affinity for cardiac glycosides. Whether or not a cell is affected depends on factors such as the Na⁺ pump reserve, the presence or absence of Na⁺/Ca⁺⁺ exchangers, and the role of Na⁺ or Ca⁺⁺ in its function.

Neurons of the autonomic nervous system are particularly sensitive to glycoside-induced Na⁺ pump inhibition, probably because of an increased baroreceptor sensitivity. Cardiac glycosides increase parasympathetic discharge, as discussed previously. Stimulation of the chemoreceptor trigger zone is responsible for nausea and vomiting. At sympathetic nerve terminals, Na⁺ pump inhibition facilitates neurotransmitter release, which is responsible for the transient vasoconstriction observed after rapid intravenous administration.

Digoxin also causes vasoconstriction by increasing intracellular Ca⁺⁺ in vascular smooth muscle. In CHF, the hemodynamic response is characterized by a decreased heart rate (caused by augmented baroreceptor responsiveness), increased forearm blood flow and cardiac index, and a decreased sympathetic tone to skeletal muscle. These beneficial changes may be the result of a reduction in neurohumoral activation, which may distinguish the cardiac glycosides from other positive inotropic agents.

Diuretics

Diuretics are widely used in treatment of congestive heart failure to reduce extracellular fluid volume (see Chapter 21). Their primary parenteral use is in patients with acute heart failure for correction of volume overload. In the most severe cases, intravenous infusions of a loop diuretic (e.g., furosemide) can initiate a rapid, predictable, and sustained diuresis. The diuretic is titrated according to an estimated “dry” weight, based on optimal filling pressures and symptoms, without exacerbating symptomatic hypotension.

Sympathomimetics

Epinephrine (Epi) and NE are β receptor agonists that produce a marked positive inotropic response and have α₁ receptor agonist activity that elicits peripheral vasoconstriction (see Chapter 11). Both agents have limited utility in patients with severe heart failure but can offer significant inotropic support for short-term intervention (minutes to hours) in life-threatening situations while more definitive measures are initiated.

Dopamine acts on prejunctional D₂ receptors to inhibit release of NE (see Chapter 19), resulting in vasodilation. Dopamine also acts on cardiac β₁ receptors to elicit a positive inotropic action and on vascular smooth muscle to cause vasodilation, improving blood flow to renal, mesenteric, coronary, and cerebral vascular beds.

Dobutamine is a racemic mixture that activates several adrenergic receptors. It cannot interact with dopaminergic receptors but releases NE from sympathetic nerves. The resulting hemodynamic effects are dose dependent, with a positive inotropic action at low doses as a result of β₁ receptor activity. The α₁ receptor-mediated vasoconstriction is attributed to the (-) enantiomer, which is countered by the receptor-blocking actions of the (+) enantiomer.

Phosphodiesterase Inhibitors

Inamrinone and milrinone exert their positive inotropic actions by inhibiting cyclic adenosine monophosphate (cAMP) phosphodiesterase (PDE), the enzyme that hydrolyzes and inactivates cAMP. The effects of these drugs differ from those of other PDE inhibitors, such as caffeine or theophylline, in that they are selective for a particular isozyme, PDE type 3. They increase cardiac cAMP concentrations but not cyclic guanosine monophosphate concentrations. This promotes cAMP-dependent protein kinase A-mediated phosphorylation of the Ca⁺⁺ channel in the heart, enhancing Ca⁺⁺ influx and resulting in a positive inotropic action similar to that caused by catecholamines.

Vasodilators

Nitroprusside was among the earliest vasodilators to show improvement in cardiac output in patients with decompensated heart failure. Nitroprusside and other nitrovasodilators are discussed in Chapter 24. Nitroprusside reduces ventricular filling pressures by directly increasing venous compliance, resulting in a redistribution of blood from central to peripheral veins. In addition, the action of nitroprusside on the arterial side of the circulation makes it one of the most effective agents for reducing left ventricular afterload. Nitroprusside dilates the pulmonary arterioles, thereby decreasing right ventricular afterload. Reducing both preload and afterload improves myocardial energetics because of a reduction in wall stress. In contrast, nitroglycerin shows specificity for venodilation, thereby increasing venous capacitance and pooling of blood in the more dependent regions of the body.

Brain (B-type) natriuretic peptide (BNP) is secreted constitutively by ventricular myocytes in response to stretch and increased wall stress and is increased in patients with CHF. Its action is counterregulatory to many of the actions of the RAAS and SNS in heart failure. BNP binds to receptors in the vasculature, kidney, and other organs, producing potent vasodilation with rapid onset and offset of action by increasing levels of cyclic guanosine monophosphate. Nesiritide is recombinant human BNP approved for treatment of acute decompensated CHF. Its acute hemodynamic effects include a reduction of right atrial, pulmonary artery, and pulmonary capillary wedge pressures, as well as systemic and pulmonary vascular resistances, causing an indirect increase in cardiac output and diuresis. Nesiritide may potentially provide symptomatic relief without the increases in mortality shown with other inotropic agents, because it does not increase cAMP and Ca⁺⁺ in cardiomyocytes.

Cardiac Glycosides

Pharmacokinetic parameters of cardiac glycosides are presented in Table 23-3. Digitoxin is rarely used in the United States, although it continues to be used in Europe. Absorption of digoxin given orally varies from 45% to 85%, and because of such wide variations, patients should be maintained on a specific brand. However, bioavailability of digitoxin is consistently high.

Digoxin is excreted mainly by the kidneys, whereas digitoxin is metabolized in the liver. Digitoxin is excreted into bile and undergoes enterohepatic cycling. The reabsorbed metabolites are cardioactive and contribute to its extended t_1/2. There are large interpatient variations in digitoxin metabolism, partly because intestinal flora exert a significant role.

Variations may be minimized by maintenance of predetermined concentrations in plasma. However, a given plasma concentration may be therapeutic in some patients and toxic in others because of differences in sensitivity caused by various factors (Table 23-4). Because monitoring the positive inotropic effect is impractical, clinical evaluation often involves the electrocardiogram. A slight (approximately 10%) increase in PR interval is not alarming; however, a greater delay in AV nodal conduction time and conduction block, or development of ventricular bigeminy or trigeminy, is a harbinger of serious toxicity.

TABLE 23–4 Factors Leading to Altered Sensitivity to Digoxin

Because digitoxin has a long t_1/2, steady-state is not achieved until 20 days after starting therapy without a loading dose; thus a loading dose is usually given. Digoxin, with its shorter t_1/2, may be administered without a loading dose; steady-state concentration is reached in 5 to 7 days when administered orally daily. Switching from maintenance doses of digoxin to digitoxin results in a temporary loss of effect, because digoxin is excreted from the body rapidly, whereas digitoxin accumulates slowly. Conversely, switching from maintenance doses of digitoxin to digoxin causes a transient overdose. Recent studies suggest that “ideal” or therapeutic digoxin serum concentrations are 1.1 to 1.2 ng/mL, whereas others favor a range of 0.5 to 1.5 ng/mL.

Phosphodiesterase Inhibitors

Inamrinone and milrinone are administered parenterally. Inamrinone is usually administered as an initial loading dose followed by careful titration. Milrinone is approximately 10 times more potent and is often preferred for short-term parenteral inotropic support in patients with severe cardiac decompensation. The elimination half-lives of inamrinone and milrinone are 2.5 hours and 30 to 60 minutes, respectively, and are approximately doubled in CHF patients.

Nesiritide

When administered intravenously to patients with CHF, as an infusion or bolus injection, nesiritide exhibits a biphasic pattern of disposition. The mean terminal elimination t_1/2 is approximately 18 minutes. Nesiritide is cleared by three mechanisms:

Neither titration of the infusion rate nor invasive hemodynamic monitoring is commonly required. Therefore patients treated with nesiritide may not require as close monitoring as may be necessary with nitrovasodilators, perhaps negating the need for intensive care unit stays.

Angiotensin-Converting Enzyme Inhibitors

ACE inhibitors now have the primary role in contemporary therapy of CHF and should be used at all costs. They must be administered in high doses, while avoiding excess hypotension. The initial dose of an ACE inhibitor must be chosen cautiously, especially in patients on diuretic therapy (most likely with intense RAAS activation); the diuretic dose must be discontinued to allow for volume expansion so as not to precipitate excessive hypotension. If the patient cannot tolerate the ACE inhibitor because of severe coughing unrelated to CHF, changing to an ARB would be the next choice.

Several clinical trials have reported significant reductions in mortality in patients receiving ACE inhibitors, even when added to the standard regimen of diuretics and digoxin. The central role of the RAAS in development and progression of cardiovascular disease, and, in particular, CHF, is well established. Short-term activation in heart failure improves cardiac output through fluid and Na⁺ retention (increased preload), whereas long-term activation results in vasoconstriction, increased afterload, and decreased cardiac output.

ACE inhibitors reduce the progression of left ventricular dysfunction in chronic heart failure and in patients recovering from an acute myocardial infarction. ACE inhibitors may also reduce the risk of acute coronary ischemic attacks. Long-term studies indicate that ACE inhibitors increase survival. An advantage in the relief of the symptoms of CHF is that they conserve K⁺ by lowering aldosterone secretion, ruling out the need for K⁺ supplementation.

It is currently unclear whether the beneficial effects of ACE inhibitors are solely the result of their hemodynamic actions, whether a reduction in the concentration of angiotensin II increases the concentrations of bradykinin or NO, or whether an inhibition of the SNS plays a significant role.

β Adrenergic Receptor Blocking Drugs

The β receptor blockers that have been shown to be effective in treatment of heart failure include those that selectively block β₁ receptors, such as metoprolol, and those that block both α₁, β₁, and β₂ receptors, such as carvedilol. Their properties are discussed in Chapter 11.

In practice, β receptor blockers are almost always used together with ACE inhibitors (and usually with digoxin). Patients need not be taking high doses of ACE inhibitors before being considered for treatment with β receptor blockers. In patients taking low doses of an ACE inhibitor, addition of a β receptor antagonist produces a greater improvement in symptoms and reduces the risk of death. The β receptor blockers should not be prescribed without diuretics in patients with a history of fluid retention, because diuretics are needed to maintain Na⁺ balance and prevent development of fluid retention that can accompany β receptor blocker therapy. Doses should be increased gradually, until side effects associated with lower doses have disappeared. Clinical trials show that 85% of patients could tolerate short-and long-term treatment with β receptor blockers.

The β receptor blockers should be prescribed to all patients with stable heart failure caused by left ventricular systolic dysfunction, unless they have a contraindication to their use or cannot tolerate treatment with these drugs. Initial doses are typically much lower than those required for hypertension and are gradually increased over time for maximal therapeutic effectiveness. Because of favorable effects on survival, treatment with β receptor blockers should not be delayed until the patient is found to be resistant to treatment with other drugs. Although it is commonly believed (incorrectly) that patients with mild symptoms or who appear clinically stable do not require additional treatment, such patients are at high risk for morbidity and mortality and are likely to deteriorate over the next year even if treated with digoxin, diuretics, and ACE inhibitors. Therefore patients with mild symptoms should also receive β receptor blockers to reduce further risk.

In summary, the β receptor antagonists are indicated in stable patients with chronic systolic heart failure and mild to moderate symptoms in combination with ACE inhibitors, diuretics, and digoxin. Therapy should be initiated slowly over several weeks with close follow-up.

Angiotensin Receptor Blockers

Administered with or without an ACE inhibitor, ARBs have been shown to increase left ventricular ejection fraction and reduce end-systolic and end-diastolic volumes at peak exercise in patients with heart failure. Clinical trials have shown that ARBs reduce morbidity and mortality in patients with heart failure. By acting at the receptor level, ARBs provide more complete blockade of the RAAS than ACE inhibitors, because angiotensin II may be formed by alternative enzymes, as discussed. These alternative pathways appear to be important, because plasma levels of angiotensin II return to pretreatment levels in some patients who receive long-term treatment with an ACE inhibitor and increase after exercise in healthy volunteers despite effective ACE inhibition. Unlike ACE inhibitors, ARBs do not activate the SNS. Potential favorable effects of ARB therapy may also be due to continued activation of AT₂ receptors, which may mediate desirable effects of vasodilation, antiproliferative effects, cell differentiation, and tissue repair. Some evidence suggests that combining an ARB with an ACE inhibitor may result in greater effects than higher doses of either drug alone.

Aldosterone Antagonists

Aldosterone concentrations are elevated as much as 20-fold in patients with heart failure. Aldosterone promotes Na⁺ retention, Mg⁺⁺ and K⁺ loss, sympathetic activation, parasympathetic inhibition, myocardial and vascular fibrosis, baroreceptor dysfunction, impaired arterial compliance, and vascular damage. Both spironolactone and eplerenone are reported to reduce mortality in patients with moderate or severe heart failure who are otherwise optimally treated. Current guidelines recommend using them in patients with severe symptoms, preserved renal function, and normal K⁺ levels. Plasma K⁺ must be monitored carefully, and caution should be exercised in patients taking K⁺ supplements or using K⁺ sparing diuretics because of the risk of hyperkalemia.

Clinical trials of spironolactone in patients with moderate or severe heart failure receiving an ACE inhibitor and a loop diuretic were discontinued after 24 months because of the significant benefits of spironolactone, including reductions in mortality (30%) and in hospitalization for worsening heart failure (35%), and improvement in symptoms. The benefit was not primarily diuretic but probably related to interference with aldosterone-mediated myocardial fibrosis and improved endothelial function.

Cardiac Glycosides

Inhibition of the Na⁺,K⁺-ATPase of the myocardial sarcolemma is responsible for both the positive inotropic and toxic effects of cardiac glycosides. A moderate (20% to 40%) inhibition causes a therapeutic effect, whereas greater inhibition is toxic. Thus the therapeutic index of these agents is narrow, because a significant positive inotropic effect requires a dose that is 50% to 60% of its toxic dose.

Cardiac glycosides are the only orally effective inotropic agents approved for use in the United States. Compared with other inotropic agents, they are unique in that they exert a direct positive inotropic response in combination with an indirectly mediated bradycardia. Therefore, despite their narrow therapeutic index, they remain important inotropic agents.

Cardiac glycosides increase the force of cardiac contraction in either normal or failing hearts. Although originally thought to be effective only in patients with heart failure, they also increase force of contraction and reduce end-diastolic volume in normal hearts, which in turn decreases the force of contraction, canceling their positive inotropic effects (see Fig. 23-1, B, arrow). In the failing dilated heart, the increased force of contraction and decrease in end-diastolic volume make the heart’s operation more nearly normal (see Fig. 23-1, C, arrow). Therefore, despite direct positive inotropic effects on both failing and nonfailing hearts, hemodynamic improvements are obtained only in the failing heart.

Because of autoregulatory mechanisms, a reduced force of cardiac contraction that lowers blood pressure triggers activation of the SNS and RAAS. The volume of circulating blood also may increase, which may result in decreased perfusion of certain organs. The primary beneficial effect of cardiac glycosides is a reversal of these changes and improvement in tissue perfusion.

Digoxin is especially useful in CHF patients with atrial fibrillation, because it slows ventricular rate, allowing for improved filling and increasing ejection fraction or stroke volume. The net result is a reduced need for heightened sympathetic tone. This unique property makes digoxin useful in CHF patients in sinus rhythm. An additional benefit is a reduction in ventricular size in the failing heart, reducing ventricular wall tension, an important determinant of O₂ consumption. This is beneficial in patients with CHF caused by ischemic heart disease. In patients with chronic CHF and abnormal systolic function, digoxin in combination with diuretics and ACE inhibitors reduces the frequency of hospitalizations and overall mortality.

Cardiac glycosides are useful in treatment of patients with chronic atrial fibrillation with a rapid ventricular response. The goal is to reduce the number of impulses from gaining access to the ventricular conducting system, thus allowing for control of ventricular rate. Other drugs (e.g., adenosine, Ca⁺⁺ channel blockers, β receptor blockers) would be additive in increasing AV nodal refractory period. Because the indirect effects of cardiac glycosides on atria lead to a decrease in effective refractory period and an increase in conduction velocity, use of digoxin is contraindicated in patients with Wolff-Parkinson-White syndrome (preexcitation) and atrial fibrillation. In such cases the number of impulses traversing the bypass tract would increase and lead to increased ventricular rate, with the potential for ventricular fibrillation.

Although digitalis glycosides have been used for more than 200 years, their narrow margins of safety and limited ability to increase ventricular function in certain clinical settings are problematic. Moreover, they cannot arrest the progression of pathological changes causing heart failure and do not prolong life in patients with CHF. However, sufficient data indicate that CHF patients maintained on digoxin experience a deterioration in cardiac function when digoxin is withdrawn, and benefit when treatment is resumed.

Diuretics

The diuretics are discussed in Chapter 21. Prospective clinical trial data do not exist for evaluating their overall efficacy on mortality in patients with heart failure. However, there is little doubt that they are useful and necessary adjuncts for relief of CHF symptoms resulting from Na⁺ and H₂O retention in patients with acute or chronic cardiac decompensation.

Parenteral administration of diuretics is useful in treating acute heart failure because they reduce circulatory congestion and pulmonary and peripheral edemas. A reduction in atrial and ventricular diastolic pressure relieves stress on the ventricular wall and promotes subendocardial perfusion. Loop diuretics and thiazides are most commonly used in patients with CHF.

The renal response to parenteral loop diuretics depends upon the peak serum concentration achieved in the renal glomeruli. A low cardiac output and an increased volume of distribution can adversely alter the anticipated response by limiting the concentration at its target site. Thus failure to achieve a response (diuretic resistance) may be due to poor renal perfusion and inadequate drug delivery. The latter may be corrected by concomitant administration of low doses of dopamine to improve renal blood flow. Although parenteral administration of loop diuretics does not directly increase myocardial contractility, there is a beneficial hemodynamic response secondary to venous dilation or increase in venous capacitance, which reduces left ventricular preload. Excessive diuresis and an excessive reduction in preload should be avoided, because the decompensated heart relies upon an expanded end-diastolic volume, which serves as a compensatory mechanism for increasing stroke volume. The dosing regimen must balance the optimal relief of edema and excess loss of fluid volume while avoiding disturbances in serum electrolytes and induction of prerenal azothermia.

Unfortunately, in more advanced CHF, the use of a single diuretic may have limited efficacy, and combination therapy may be required. With “diuretic resistance,” it is common to use a combination of a loop and a distal tubular diuretic (see Chapter 21).

Sympathomimetics

Despite the fact that NE and Epi increase cardiac contractility and systemic blood pressure, the major drawbacks to their use in acute heart failure is the intense increase in peripheral vasoconstriction and increase in left ventricular afterload, thus further impairing cardiac output in an already failing heart. The peripheral vasoconstrictor effects lead to impaired tissue perfusion especially to heart, kidney, and splanchnic regions. Furthermore activation of cardiac β₁ receptors may result in an increased O₂ demand, leading to development of relative myocardial ischemia and potentially lethal cardiac arrhythmias.

The combination of selective vasodilator effects and β₁ receptor activation by dopamine make it attractive for situations in which blood pressure is low and renal perfusion is poor, as in cardiogenic, traumatic, or hypovolemic shock. As discussed, its renal vasodilator action is additive to the effects of furosemide, making their combined use an important adjunctive intervention in patients with acute cardiac decompensation and volume overload, or in “diuretic-resistant” patients. Dopamine is given intravenously because of its short t_1/2 and rapid metabolism.

Dobutamine is useful in patients with low cardiac output and an increased left ventricular end-diastolic pressure who are not hypotensive. Under such circumstances the positive inotropic action (β₁ receptor-mediated) and the ability to reduce left ventricular afterload (β₂ receptor-mediated vasodilation) would augment stroke volume and improve organ perfusion with little increase in heart rate. However, long-term use of dobutamine is limited by the development of tolerance. In patients with acute decompensated heart failure who are in need of short-term inotropic support, dobutamine is preferred over dopamine. At higher doses dobutamine will increase systemic arterial pressure and increase ventricular afterload.

Heart rate may increase during dobutamine administration. This is of particular concern in patients with atrial fibrillation, where β receptor activation at the AV node will increase atrial impulses to the ventricular conducting system. The short t_1/2 of dobutamine is advantageous when unexpected hypotension, tachycardia, or tachyarrhythmia results. As with all β receptor agonists, dobutamine will be ineffective in patients being treated with β receptor blockers.

Phosphodiesterase Inhibitors

Inamrinone and milrinone were introduced as oral agents for treatment of patients with chronic CHF, although they are now used primarily parenterally for management of acute heart failure. They have direct positive inotropic effects and increase the rate of myocardial relaxation. They also cause a balanced arterial and venous vasodilation, leading to decreased arterial and pulmonary vascular resistance. The result is an increased cardiac output caused by an increased myocardial contractility and a decreased ventricular afterload.

Both inamrinone and milrinone are effective in patients receiving β receptor blockers. They may therefore serve as a “bridge to β receptor blockade” for long-term treatment of patients with severe refractory heart failure who are unable to tolerate β receptor blockers in the absence of added inotropic support. The increase in stroke volume and cardiac output observed with inamrinone and milrinone are due mainly to peripheral vasodilation and a decrease in left ventricular afterload. There is significant variability in the degree to which cardiac output increases and systemic vascular resistance decreases. Clinically significant hypotension has occurred with milrinone.

Vasodilators

Vasodilators used for acute or chronic treatment of heart failure should be given in doses that reduce peripheral resistance but do not cause a sharp decrease in blood pressure, that is, in doses at which most blood pressure effects are compensated for by homeostatic mechanisms. These drugs relax venous and arterial smooth muscle, thereby reducing resistance to ventricular ejection and increasing the capacity of the venous reservoir. This causes relief of symptoms and an increase in exercise tolerance in patients with a dilated ventricle. These changes can be achieved acutely by intravenous nitroprusside, nitroglycerin, and nesiritide, or by chronic administration of hydralazine together with isosorbide dinitrate or an ACE inhibitor (see Chapter 24).

The major hemodynamic effect of nitroglycerin is a reduction in preload and a decrease in left ventricular end-diastolic pressure. However, in the presence of increased peripheral vascular resistance and with relatively high doses administered intravenously, nitroglycerin also elicits a vasodilator effect on the arterial circulation. Nitroglycerin infusion is used for patients with acute ischemic syndromes or in patients with acute decompensated heart failure caused by ischemic heart disease. Tolerance occurs and is clinically important with prolonged administration.

Although the renal effects of BNP on the kidney and its ability to unload the heart are well documented, some data suggest that it also has direct actions on cardiac fibroblasts. BNP is found in cardiac fibroblasts and inhibits de novo collagen synthesis and increases expression of specific matrix metalloproteinases. It may therefore be beneficial in controlling synthesis and degradation of collagen deposition after myocardial injury. Also, chronic administration of BNP suppresses aldosterone secretion, despite a natriuretic response. Thus BNP works through both renal mechanisms and suppression of the profibrotic action of aldosterone.

Intravenous infusion of nesiritide (recombinant BNP) in patients with CHF results in beneficial hemodynamic actions, including arterial and venous dilation, enhanced Na⁺ excretion, and suppression of the RAAS and SNS. Nesiritide alleviates the symptoms of acute decompensated heart failure and is useful in augmenting the effects of loop diuretics in patients who fail to respond with an adequate diuresis.

Cardiac Glycosides

Digoxin toxicity remains an important clinical problem that demands vigilance for the early recognition of disturbances of cardiac impulse formation and conduction abnormalities, along with more subtle signs related to the central nervous and gastrointestinal systems (see Clinical Problems Box). Toxic effects of digoxin may occur at any serum concentration due to factors that affect sensitivity (see Table 23-4) because of its low therapeutic index.

Several factors affect the sensitivity of the heart to cardiac glycosides. Binding to the Na⁺,K⁺-ATPase is slow and enhanced by high intracellular Na⁺ and low extracellular K⁺. Thus its pharmacological and toxic effects are greater in hypokalemic patients. K⁺ depleting diuretics are a major contributing factor to digoxin toxicity. Ca⁺⁺, if administered rapidly intravenously, may produce serious arrhythmias in patients treated with cardiac glycosides. Tachycardia, which increases Na⁺ influx, also enhances their actions. Larger doses are used in newborn and young infants than in adults because of the low sensitivity of infant heart muscle to glycosides.

When cardiac muscle is exposed to toxic concentrations of a glycoside, Na⁺ pump inhibition and cellular Ca⁺⁺ loading become excessive. The cytoplasmic membrane becomes unstable for a short time immediately after each membrane repolarization. In normal ventricular muscle cells, membrane depolarization is followed by repolarization, so that the membrane potential reaches approximately -90 mV and remains there until the next wave of depolarization (Fig. 23-4, A). In digoxin toxicity, however, the membrane becomes more permeable to Na⁺, Ca⁺⁺, and K⁺ immediately after repolarization. Movements of Na⁺ and Ca⁺⁺ are particularly prominent, because these ions are driven by both chemical and electrical gradients.

FIGURE 23–4 Changes in cardiac action potentials caused by subtoxic and toxic doses of cardiac glycosides. A, Typical action potential recordings from cardiac Purkinje fiber cells. Toxic doses produce oscillatory afterdepolarizations (B) and ventricular tachycardia (C).

In “mild” toxicity the transient inward current subsides, causing the transmembrane potential to return to its resting level. This may be repeated several times, causing oscillatory delayed afterpotentials (Fig. 23-4, B). These small oscillatory delayed afterpotentials are most readily observed in cardiac Purkinje fibers and do not propagate beyond the individual cell. When their magnitude increases in advanced digoxin toxicity, the threshold potential is reached, causing the cell to be depolarized (i.e., to trigger action potentials; see Fig. 23-4, C). Such triggered action potentials propagate from Purkinje fiber cells to ventricular muscle, causing the muscle to contract repetitively and no longer be under the control of the SA node. This may lead to life-threatening ventricular tachycardia, fibrillation, or both.

Thyroid hormone administration to a digitalized, hypothyroid patient may increase its dose requirement. Simultaneous use of digoxin and sympathomimetics increases the risk of cardiac arrhythmias by enhancing the formation of delayed afterdepolarizations. Administration of succinylcholine results in sudden release of K⁺ from skeletal muscle and may cause an increase in AV block in digitalized patients. Although β receptor blockers or Ca⁺⁺ channel blockers and digoxin may be useful in combination to control atrial fibrillation, their additive effects on AV node conduction can result in advanced or complete heart block. The reduction in plasma K⁺ after administration of insulin may be associated with development of cardiac arrhythmias in patients receiving a cardiac glycoside. Caution should be exercised when combining digoxin with any drug that may cause a significant deterioration in renal function, because a decline in glomerular filtration or tubular secretion may impair its excretion.

Pharmacokinetic interactions involving digoxin are considerable. Quinidine should not be used to treat digoxin-induced arrhythmias because it increases plasma digoxin concentration, apparent volume of distribution, and renal clearance. Other drugs also interact with digoxin. Significant increases in plasma digoxin may occur with verapamil, nifedipine, amiodarone, or quinine, but only at high doses.

Ethacrynic acid, furosemide, and thiazide diuretics increase the therapeutic and toxic effects of digoxin and reduce its therapeutic index by causing K⁺ depletion. Propranolol may augment bradycardia, whereas barbiturates, phenytoin, and phenylbutazone can enhance metabolism. Cholestyramine combines with digitoxin in the intestine and enhances its elimination.

Primary signs of digoxin toxicity include arrhythmias caused by suppression of AV nodal conduction. The first action taken upon suspecting digitalis toxicity is to discontinue administration until the adverse reaction resolves or is determined to be unrelated to the drug. Ventricular premature contractions triggered by oscillatory afterpotentials that originate in Purkinje fibers may also be superimposed. These arrhythmias may be converted to normal sinus rhythm by K⁺ when the plasma K⁺ concentration is low or within the normal range. K⁺ is often effective against glycoside-induced arrhythmias because it:

When the plasma K⁺ concentration is high, antiarrhythmic drugs such as lidocaine, procainamide, or propranolol can be used. Although phenytoin is reported to be useful in treating arrhythmias, there have been several instances of sudden death in patients when phenytoin has been administered to treat glycoside overdose. The most dramatic treatment for digoxin toxicity is a specific antibody raised against digoxin, which is administered intravenously and binds serum digoxin. The complex is excreted rapidly by the kidney.

In the more severe situation where the patient exhibits a disturbance in cardiac rhythm, additional therapy may be required. In the presence of symptomatic bradyarrhythmia or heart block, consideration should be given to the reversal of toxicity with a digoxin antibody, the use of atropine, or placement of a temporary cardiac pacemaker. However, asymptomatic bradycardia or heart block related to digoxin may require only temporary withdrawal of the drug and cardiac monitoring of the patient.

In the presence of a more severe and potentially life-threatening arrhythmia, such as bidirectional ventricular tachycardia, consideration should be given to the correction of electrolyte disorders, particularly if hypokalemia or hypomagnesemia are present. The antidigoxin antibody may be used to reverse potentially life-threatening ventricular arrhythmias.

Other Drugs for Congestive Heart Failure

The side effects of β receptor blockers are discussed in detail in Chapter 11. Adverse effects are generally an extension of their therapeutic actions and include cardiac decompensation, bradycardia, hypoglycemia, and cold extremities. Initiation of treatment with a β receptor blocker has produced four types of adverse reactions that require careful attention and management:

ACE inhibitors often cause cough and, less commonly, development of angioneurotic edema. Both ACE inhibitors and ARBs should be discontinued before the second trimester of pregnancy because of their potential for teratogenic effects. Hypotension, oliguria, progressive azothermia, and hyperkalemia are not uncommon.

Contraindications for use of ACE inhibitors include bilateral renal artery stenosis and known allergies. High serum creatinine is a contraindication for use of ACE inhibitors or ARBs, as is the presence of hyperkalemia that will worsen on drug therapy. Side effects of ACE inhibitors and ARBs are discussed in more detail in Chapter 20.

Adverse effects associated with aldosterone antagonists include hyperkalemia, agranulocytosis, anaphylaxis, hepatotoxicity, and renal failure (see Chapter 39). Spironolactone has the added features of inducing gynecomastia, sexual dysfunction, and menstrual irregularities. Severe adverse effects associated with eplerenone include severe arrhythmias, life-threatening myocardial infarction, and myocardial ischemia (angina).

The sympathomimetics Epi and NE may cause restlessness, headache, tremor, and cardiac palpitations, as well as cerebral hemorrhage and cardiac arrhythmias (see Chapter 11). They should be used with caution in patients receiving nonselective β receptor blockers, because their unopposed actions on vascular α₁ receptors can cause an acute hypotensive crisis and possible cerebral hemorrhage. The adverse effects of dopamine and dobutamine are similar and attributable to excessive sympathomimetic activity related to overdose (see Chapter 11).

Serious adverse effects attributable to the PDE inhibitor inamrinone include ventricular arrhythmias, hypotension, and the potential for development of thrombocytopenia (10%) on prolonged administration. Long-term clinical trials of milrinone and inamrinone were associated with significant adverse effects and increased mortality in patients with heart failure. Currently, intravenous formulations of inamrinone and milrinone are approved for short-term support in patients with acute cardiac decompensation who are unresponsive to other drugs, such as diuretics or digoxin.

Adverse effects of nitrovasodilators, most commonly hypotension, are discussed in Chapter 24. Aggressive treatment with nitroprusside may result in a precipitous fall in left ventricular end-diastolic pressure, marked hypotension, and myocardial ischemia. The accompanying

pulmonary vasodilation may lead to an increased ventilation-perfusion mismatch and hypoxia. Nitroprusside is contraindicated in patients with severe obstructive valvular heart disease (aortic, mitral, or pulmonic stenosis or obstructive cardiomyopathy).

Hypotension, occasionally accompanied by bradycardia, is the major side effect associated with the administration of nesiritide, which is usually well tolerated in the supine position. The potential for hypotension is increased with concomitant administration of other drugs capable of lowering blood pressure.