Chapter 28: Pharmacotherapy of Congestive Heart Failure

Congestive heart failure (CHF) is responsible for more than half a million deaths annually in the U.S., carries a 1-year mortality rate of more than 50% in patients with advanced forms of the condition, and is responsible for nearly $27 billion annually in healthcare costs (Binanay et al., 2005). Fortunately, substantive advances in CHF pharmacotherapy have altered clinical practice by shifting the paradigm of its management from exclusively symptom palliation to modification of disease progression and, in some cases, an expectation of prolonged survival.

In the past, drug therapies targeted the endpoints of this syndrome: volume overload (congestion) and myocardial dysfunction (pump failure). As a consequence, diuretics and cardiac glycosides dominated the medical management of CHF for more than 40 years. These drugs remain effective for symptom relief and in stabilizing patients with hemodynamic decompensation but do not improve long-term survival. The contemporary focus of CHF as a disorder of circulatory hemodynamics, pathologic cardiac remodeling, and increased arrhythmogenic instability has translated into the development of novel pharmacotherapies that reduce CHF-associated morbidity and mortality rates. Before discussing the clinical pharmacology of CHF in specific, it is useful to establish a pathophysiologic framework through which its treatment is approached.

Defining Congestive Heart Failure. The onset and progression of clinically evident CHF from left ventricular (LV) systolic dysfunction follows a pathophysiologic sequence in response to an initial insult to myocardial dysfunction. A reduction in forward cardiac output leads to expanded activation of the sympathetic nervous system and the renin–angiotensin–aldosterone axis that, together, maintain perfusion of vital organs by increasing LV preload, stimulating myocardial contractility, and increasing arterial tone. Acutely, these mechanisms sustain cardiac output by allowing the heart to operate at elevated end-diastolic volumes, while peripheral vasoconstriction promotes regional redistribution of the cardiac output to the central nervous system, coronary, and renal vascular beds.

Unfortunately, however, these compensatory mechanisms over time propagate disease progression. Intravascular volume expansion increases diastolic and systolic wall stress that disrupts myocardial energetics and causes pathologic LV hypertrophy. By increasing LV afterload, peripheral arterial vasoconstriction also adversely affects diastolic ventricular wall stress, thereby increasing myocardial O₂ demand. Finally, neurohumoral effectors such as norepinephrine (NE) and angiotensin II (AngII) are associated with myocyte apoptosis, abnormal myocyte gene expression, and pathologic changes in the extracellular matrix that increase LV stiffness (Villarreal, 2005).

In clinical practice, the term CHF describes a final common pathway for the expression of myocardial dysfunction. While some emphasize the clinical distinction between systolic versus diastolic heart failure, many patients demonstrate dysfunction in both contractile performance and ventricular relaxation/filling. Indeed, these physiologic processes are interrelated; for example, the rate and duration of LV diastolic filling are directly influenced by impairment in systolic contractile performance. Despite inherent difficulties in creating a singular description to encompass the diverse pathophysiologic mechanisms that result in CHF (e.g., myocardial infarction (MI), viral myocarditis), the following definitions are useful for establishing a conceptual framework to describe this clinical syndrome:

Congestive heart failure is the pathophysiologic state in which the heart is unable to pump blood at a rate commensurate with the requirements of metabolizing tissues, or can do so only from an elevated filling pressure (Braunwald and Bristow, 2000).

Heart failure is a complex of symptoms—fatigue, shortness of breath, and congestion—that are related to the inadequate perfusion of tissue during exertion and often to the retention of fluid. Its primary cause is an impairment of the heart's ability to fill or empty the left ventricle properly (Cohn, 1996).

From these definitions, one may consider CHF as a condition in which failure of the heart to provide adequate forward output at normal end-diastolic filling pressures results in a clinical syndrome of decreased exercise tolerance with pulmonary and systemic venous congestion. Numerous cardiovascular comorbidities are associated with CHF, including coronary artery disease, MI, and sudden cardiac death.

The abnormalities of myocardial structure and function that characterize CHF are often irreversible. These changes narrow the end-diastolic volume range that is compatible with normal cardiac function. Although CHF is predominately a chronic disease, subtle changes to an individual's hemodynamic status (e.g., increased circulating volume from high dietary sodium intake, increased systemic blood pressure from medication nonadherence) often provoke an acute clinical decompensation.

Not surprisingly, therefore, CHF therapy has for many years utilized diuretics to control volume overload and subsequent worsening in LV function. Other proven pharmacotherapies target ventricular wall stress, the renin–angiotensin–aldosterone axis, and the sympathetic nervous system to decrease pathologic ventricular remodeling, attenuate disease progression, and improve survival in selected patients with severe CHF and low LV ejection fraction. Figure 28–1 provides an overview of the sites of action for major drug classes commonly used in clinical practice. Note that these therapies largely improve cardiac hemodynamics and function through preload reduction, afterload reduction, and enhancement of inotropy (i.e., myocardial contractility). In addition, pharmacotherapeutics that target peripheral and coronary vascular function are receiving increased attention as potential participants in the comprehensive management of CHF patients.

Diuretics

Diuretics remain central in the pharmacologic management of congestive symptoms in patients with CHF. The pharmacologic properties of these agents are presented in detail in Chapter 25. Their importance (and frequency of use) in CHF management reflects the deleterious downstream effects of volume expansion on increased LV end-diastolic volume, an intermediate step in the development of elevated right-heart pressure, pulmonary venous congestion, and peripheral edema (Hillege et al., 2000).

Diuretics reduce extracellular fluid volume and ventricular filling pressure (or "preload"). Because CHF patients often operate on a "plateau" phase of the Frank-Starling curve (Figure 28–2), incremental preload reduction occurs under these conditions without a reduction in cardiac output. Sustained natriuresis and/or a rapid decline in intravascular volume, however, may "push" one's profile leftward on the Frank-Starling curve, resulting in an unwanted decrease in cardiac output. In this way, excessive diuresis is counterproductive secondary to reciprocal neurohormonal overactivation from volume depletion (McCurley et al., 2004). For this reason, it is preferable to avoid diuretics in patients with asymptomatic LV dysfunction and to only administer the minimal dose required to maintain euvolemia in those patients symptomatic from hypervolemia. Despite the efficacy of loop or thiazide diuretics in controlling congestive symptoms and improving exercise capacity, their use is not associated with a reduction in CHF mortality.

Dietary Sodium Restriction. All patients with clinically significant LV dysfunction, regardless of symptom status, should be advised to limit dietary sodium intake to 2-3 g/day. More stringent salt restriction is seldom necessary and may be counterproductive, as it can lead to hyponatremia, hypokalemia, and hypochloremic metabolic alkalosis when combined with loop diuretic administration.

Loop Diuretics. Of the loop diuretics currently available, furosemide (lasix, others), bumetanide (bumex, others), and torsemide (demadex, others) are widely used in the treatment of CHF. Due to the increased risk of ototoxicity, ethacrynic acid (edecrin) is recommended only for patients with sulfonamides allergies or who are intolerant to alternative options.

Loop diuretics inhibit a specific ion transport protein, the Na⁺-K⁺-2Cl^– symporter on the apical membrane of renal epithelial cells in the ascending limb of the loop of Henle to increase Na⁺ and fluid delivery to distal nephron segments (see Chapter 25). These drugs also enhance K⁺ secretion, particularly in the presence of elevated aldosterone levels, as is typical in CHF.

Thiazide Diuretics. Monotherapy with thiazide diuretics (diuril, hydrodiuril, others) has a limited role in CHF. However, combination therapy with loop diuretics is often effective in those refractory to loop diuretics alone. Thiazide diuretics act on the Na⁺Cl⁻ cotransporter in the distal convoluted tubule (see Chapter 25) and are associated with a greater degree K⁺ wasting per fluid volume reduction when compared to loop diuretics (Gottlieb, 2004).

K⁺-Sparing Diuretics. K⁺-sparing diuretics (see Chapter 25) inhibit apical membrane Na⁺-conductance channels in renal epithelial cells (e.g., amiloride, triamterene) or are mineralocorticoid (e.g., aldosterone) receptor antagonists (e.g., canrenone [not commercially available in the U.S.], spironolactone, and eplerenone). Collectively, these agents are weak diuretics, but have historically been used to achieve volume reduction with limited K⁺ and Mg²⁺ wasting. The beneficial role of aldosterone receptor blockers on survival in CHF is discussed later.

Diuretics in Clinical Practice. The majority of CHF patients will require chronic administration of a loop diuretic to maintain euvolemia. In patients with clinically evident fluid retention, furosemide typically is started at a dose of 40 mg once or twice daily, and the dosage is increased until an adequate diuresis is achieved. A larger initial dose may be necessary in patients with advanced CHF and azotemia. Serum electrolytes and renal function are monitored frequently in these patients or in those for whom a rapid diuresis is necessary. If present, hypokalemia from therapy may be corrected by oral or intravenous K⁺ supplementation or by the addition of a K⁺-sparing diuretic. When appropriate, diuretics are decreased to the minimum effective concentration for maintaining euvolemia.

Diuretics in the Decompensated Patient. In patients with decompensated CHF warranting hospital admission, repetitive intravenously administered boluses or a constant infusion titrated to achieve a desired response may be needed to provide expeditious (and reliable) diuresis (Dormans et al., 1996). One advantage to intravenous infusion is that sustained natriuresis is achieved as a consequence of consistently elevated drug levels within the lumen of the renal tubules. In addition, the risk of ototoxicity is reduced by continuous infusion when compared to repetitive, intermittent intravenous dosing (Lahav et al., 1992).

Diuretic Resistance. As mentioned earlier, a compensatory increase in renal tubular Na⁺ reabsorption may prevent effective diuresis when dosed daily; as a result, reduction of diuretic dosing intervals may be warranted. In advanced CHF, invasive assessment of intracardiac filling pressures and cardiac output may be required to distinguish between low intravascular volume from aggressive diuresis versus low cardiac output states, although both states are aligned with lower diuretic delivery and drug efficacy. Furthermore, edema, decreased bowel-wall motility, and reduced splanchnic blood flow impair absorption and may delay or attenuate peak diuretic effect.

Atherosclerotic renal artery disease is associated with reduced renal perfusion pressures to levels below that necessary for adequate drug delivery. In patients with reduced renal arterial perfusion pressure (e.g., renal artery stenosis or low cardiac output), AngII-mediated efferent glomerular arteriole tone is important for preservation of normal glomerular filtration pressure. Angiotensin-converting enzyme (ACE) inhibitors or AT₁ receptor antagonists in combination with loop diuretics may, therefore, be met with a decline in creatinine clearance that is associated with low diuretic delivery and decreased drug efficacy (Ellison, 1999). Other common causes of diuretic resistance are listed in Table 28–1.

Adenosine A₁ Receptor Antagonists. Adenosine A₁ receptor antagonists may provide a renal protective therapeutic strategy for enhanced volume loss in decompensated CHF. Adenosine is secreted from the macula densa in the renal arteriole in response to diuretic-induced increases in Na⁺ and Cl⁻ tubular flow concentrations. This results in increased Na⁺ resorption, a volume-loss counterregulatory mechanism (see Chapter 26). Na⁺ reabsorption, in addition to adenosine-induced renal arteriole vasoconstriction, appears responsible (in part) for the development of complications common to the use of diuretics in decompensated CHF patients, particularly prerenal azotemia. The role of adenosine in the macula densa and juxtaglomerular (granular) cells (see Figure 26–2) suggests other effects of A₁ antagonists on the renin-angiotensin system.

Aldosterone Antagonistsand Clinical Outcome

LV systolic dysfunction deceases renal blood flow and results in overactivation of the renin–angiotensin– aldosterone axis and may increase circulating plasma aldosterone levels in CHF to 20-fold above normal (Figure 28–3). The pathophysiologic effects of hyperaldosteronemia are diverse (Table 28–2) and extend beyond Na⁺ and fluid retention; importantly, however, the precise mechanism by which aldosterone receptor blockade improves outcome in CHF remains unresolved (Weber, 2004).

Vasodilators

The rationale for oral vasodilator drugs in the pharmacotherapy of CHF derives from experience with parenterally administered phentolamine and nitroprusside in patients with advanced disease and elevated systemic vascular resistance (Cohn and Franciosa, 1977). Although numerous vasodilators have since been developed that improve CHF symptoms, only the hydralazine–isosorbide dinitrate combination, ACE inhibitors, and AT₁ receptor blockers (ARBs) demonstrably improve survival. The therapeutic use of vasodilators in the treatment of hypertension and myocardial ischemia is considered in detail in Chapter 27. This chapter will focus on the uses for some of these same vasodilator drugs in the treatment of CHF. Table 28–3 summarizes properties of vasodilators commonly used to treat CHF.

Nitrovasodilators. Nitrovasodilators are nitric oxide (NO) donors that activate soluble guanylate cyclase in vascular smooth muscle cells, leading to vasodilation. The mechanism underlying the variable response profiles to nitrovasodilators in different vascular beds remains controversial; e.g., nitroglycerin preferentially induces epicardial coronary artery vasodilation. Furthermore, the mechanisms by which nitrovasodilators are converted to their active forms in vivo depend on the particular agent. Unlike nitroprusside, which is converted to NO^• by cellular reducing agents such as glutathione, nitroglycerin and other organic nitrates undergo a more complex enzymatic biotransformation to NO^• or bioactive S-nitrosothiols. The activities of specific enzyme(s) and cofactor(s) required for this biotransformation appear to differ by target organ and even by different vasculature beds within a particular organ (Münzel et al., 2005).

Organic Nitrates. As discussed in Chapter 27, organic nitrates are available in a number of formulations that include rapid-acting nitroglycerin tablets or spray for sublingual administration, short-acting oral agents such as isosorbide dinitrate (isordil, others), long-acting oral agents such as isosorbide mononitrate (ismo, others), topical preparations such as nitroglycerin ointment and transdermal patches, and intravenous nitroglycerin. These preparations are relatively safe and effective; specifically, their principal action in CHF is reducing LV filling pressure. This occurs, in part, by augmentation of peripheral venous capacitance that results in preload reduction. Additional effects of organic nitrates include pulmonary and systemic vascular resistance reduction, particularly at higher doses, and epicardial coronary artery vasodilation for which systolic and diastolic ventricular function is enhanced by increased coronary blood flow. Collectively, these beneficial physiologic effects translate into improved exercise capacity and CHF-symptom reduction. However, these drugs do not substantially influence systemic vascular resistance, and pharmacologic tolerance greatly limits their utility over time. For these reasons and others, organic nitrates are not commonly used alone; rather, a number of trials have shown that when used together, selected organic nitrates increase the clinical effectiveness of other vasodilators, such as hydralazine, resulting in a sustained improvement in hemodynamics.

Nitrate tolerance. Nitrate tolerance may limit the long-term effectiveness of these drugs in the treatment of CHF. Blood nitrate levels may be permitted to fall to negligible levels for at least 6-8 hours each day (see Chapter 27). The timing of nitrate withdrawal symptoms, if present, may be useful, however, for developing an appropriate drug-dosing schedule. Patients with recurrent orthopnea or paroxysmal nocturnal dyspnea, e.g., might benefit from nighttime nitrate use. Likewise, co-treatment with hydralazine (e.g., isosorbide dinitrate and hydralazine [BIDIL]) may decrease nitrate tolerance by an antioxidant effect that attenuates superoxide formation, thereby increasing the bioavailable NO levels (Münzel et al., 1996).

Parenteral Vasodilators

Sodium Nitroprusside. Sodium nitroprusside (nitropress, others) is a direct NO donor and potent vasodilator that is effective in reducing both ventricular filling pressure and systemic vascular resistance. The downstream beneficial physiologic effects of afterload reduction in CHF are outlined in Figure 28–4. Onset to activation of sodium nitroprusside is rapid (2-5 minutes), and the drug is quickly metabolized to NO, properties that afford easy titration to achieve the desired hemodynamic effect.

Nitroprusside is particularly effective in treating critically ill patients with CHF who have elevated systemic vascular resistance or mechanical complications that follow acute MI (e.g., mitral regurgitation or ventricular septal defect-induced left-to-right shunts). As with other vasodilators, the most common adverse side effect of nitroprusside is hypotension. In general, nitroprusside initiation in patients with severe CHF results in increased cardiac output and a parallel increase in renal blood flow, improving both glomerular filtration and diuretic effectiveness. However, excessive reduction of systemic arterial pressure may limit or prevent an increase in renal blood flow in patients with more severe LV contractile dysfunction.

Intravenous Nitroglycerin. Intravenous nitroglycerin, like nitroprusside, is a vasoactive NO donor that is commonly used in the intensive care unit setting. Unlike nitroprusside, nitroglycerin is relatively selective for venous capacitance vessels, particularly at low infusion rates. In CHF, intravenous nitroglycerin is most commonly used in the treatment of LV dysfunction due to an acute myocardial ischemia. Parenteral nitroglycerin also is used in the treatment of nonischemic cardiomyopathy when expeditious LV filling pressure reduction is desired. At higher infusion rates, this drug also may decrease systemic arterial resistance, although this effect is less predictable. Nitroglycerin therapy may be limited by headache and nitrate tolerance; tolerance may be partially offset by increasing the dosage. Administration requires the use of an infusion pump capable of controlling the rate of administration.

Hydralazine. Hydralazine is a direct vasodilator that has long been in clinical use, yet its precise mechanism of action is poorly understood. The effects of this agent are not mediated through recognized neurohumoral systems, and its mechanism of action at the cellular level in vascular smooth muscle is uncertain. Lack of mechanistic understanding not with standing, hydralazine is an effective antihypertensive drug (see Chapter 27), particularly when combined with agents that blunt compensatory increases in sympathetic tone and salt and water retention. In CHF, hydralazine reduces right and left ventricular afterload by reducing pulmonary and systemic vascular resistance. This results in an augmentation of forward stroke volume and a reduction in ventricular wall stress in systole. Hydralazine also appears to have moderate "direct" positive inotropic activity in cardiac muscle independent of its afterload-reducing effects. Hydralazine is effective in reducing renal vascular resistance and in increasing renal blood flow to a greater degree than are most other vasodilators, with the exception of ACE inhibitors. For this reason, hydralazine often is used in CHF patients with renal dysfunction intolerant of ACE-inhibitor therapy.

Targeting Neurohormonal Regulation: The Renin–Angiotensin–Aldosterone Axis and Vasopressin Antagonists

Renin–Angiotensin–Aldosterone Axis Antagonists. The renin–angiotensin–aldosterone axis plays a central role in the pathophysiology of CHF (Figure 28–3). AngII is a potent arterial vasoconstrictor and an important mediator of Na⁺ and water retention through its effects on glomerular filtration pressure and aldosterone secretion. AngII also modulates neural and adrenal medulla catecholamine release, is arrhythmogenic, promotes vascular hyperplasia and myocardial hypertrophy, and induces myocyte death. Consequently, reduction of the effects of AngII constitutes a cornerstone of CHF management (Weber, 2004).

ACE inhibitors are preferential arterial vasodilators. ACE-inhibitor–mediated decreases in LV afterload result in increased stroke volume and cardiac output; ultimately, the magnitude of these effects is associated with the observed change in mean arterial pressure. Heart rate typically is unchanged with treatment, often despite decreases in systemic arterial pressure, a response that probably is a consequence of decreased sympathetic nervous system activity from ACE inhibition.

Most clinical actions of AngII, including its deleterious effects in CHF, are mediated through the AT₁ angiotensin receptor, whereas AT₂ receptor activation appears to counterbalance the downstream biologic effects of AT₁ receptor stimulation. Owing to enhanced target specificity, AT₁ receptor antagonists more efficiently block the effects of AngII than do ACE inhibitors. In addition, the elevated level of circulating AngII that occurs secondary to AT₁ receptor blockade results in a relative increase in AT₂ receptor activation. Unlike ACE inhibitors, AT₁ blockers do not influence bradykinin metabolism (see the next section).

Angiotensin-Converting Enzyme Inhibitors. The first orally administered ACE inhibitor, captopril (capoten, others), was introduced in 1977. Since then, six additional ACE inhibitors—enalapril (vasotec, others), ramipril (altace, others), lisinopril (prinivil, zestril, others), quinapril (accupril, others), trandolapril (mavik, others) and fosinopril (monopril, others) have been FDA- approved for the treatment of CHF. Data from numerous clinical trials involving well over 100,000 patients support ACE inhibition in the management of CHF of any severity, including those with asymptomatic LV dysfunction.

ACE-Inhibitor Side Effects. Elevated bradykinin levels from ACE inhibition are associated with angioedema, a potentially life-threatening drug side effect. If this occurs, immediate and permanent cessation of all ACE inhibitors is indicated. Angioedema may occur at any time over the course of ACE inhibitor therapy. A characteristic, dry cough from the same mechanism is common; in this case, substitution of an AT₁ receptor antagonist for the ACE inhibitor often is curative. A small rise in serum K⁺ levels is common with ACE-inhibitor use. This increase may be substantial, however, in patients with renal impairment or in diabetic patients with type IV renal tubular acidosis. Mild hyperkalemia is best managed by institution of a low-potassium diet but may require drug dose adjustment. The inability to implement ACE inhibitors as a consequence of cardiorenal side effects (e.g., excessive hypotension, progressive renal insufficiency, hyperkalemia) is itself a poor prognostic indicator in the CHF patient (Kittleson et al., 2003).

ACE Inhibitors and Survival in CHF. A number of randomized, placebo-controlled clinical trials have demonstrated that ACE inhibitors improve survival in patients with CHF due to systolic dysfunction, independent of etiology. For example, ACE inhibitors prevent mortality and the development of clinically significant LV dysfunction after acute MI (Pfeffer et al., 1992; Konstam et al., 1992; St. John Sutton et al., 1994). This occurs through attenuation or prevention of increased LV end-diastolic and end-systolic volumes, and the decline in LV ejection fraction that is central to the natural history of AMI. ACE inhibitors appear to confer these benefits by preventing postinfarction-associated adverse ventricular remodeling.

AT₁Receptor Antagonists. Activation of the AT₁ receptor mediates most of the deleterious effects of AngII that are described earlier. AT₁-receptor antagonism, in turn, largely obviates AngII "escape" and substantially decreases the probability of developing bradykinin-mediated side effects associated with ACE inhibition. Importantly, however, although rare, angioedema has been reported with AT₁-receptor antagonist use. Therefore, caution is still warranted when prescribing these agents to patients with a history of ACE-inhibitor–associated angioedema.

Direct Renin Inhibitors. There is accumulating scientific and clinical evidence to suggest that maximal pharmacologic ACE inhibition alone is insufficient for optimal attenuation of AngII-induced cardiovascular dysfunction in patients with CHF. Several molecular mechanisms have been ascribed to explain this hypothesis (Abassi et al., 2009), including:

• the presence of ACE-independent pathways that facilitate AngI conversion to AngII

• the activation of ACE homologues (e.g., ACE2) that are insensitive to conventional ACE-I therapy

• the suppression of the negative feedback effect exerted by AngI on renin secretion in the kidney

For these reasons, inhibition of renin has pharmacotherapeutic objectives for enhanced suppression of AngII synthesis in CHF.

Renin-mediated conversion of angiotensinogen to AngI is the first and rate-limiting step in the biochemical cascade that generates AngII and aldosterone (see Figures 26–1 and 28–3). The first generation of orally administered direct renin inhibitors was studied in the 1980s and was designed for the treatment of hypertension, but the broader use of these agents was hampered by the drugs' suboptimal efficacy, reflecting, at least in part, their limited bioavailability (Staessen et al., 2006). Aliskiren (tekturna, rasilez) is the first orally administered direct renin inhibitor to obtain FDA approval for use in clinical practice. The pharmocokinetic advantages of aliskiren over earlier direct renin inhibitor prototypes include an increased bioavailability (2.7%) and a long plasma t_1/2 (~23 hour). Pilot clinical trials in humans established that aliskiren induces a concentration-dependent decrease in plasma renin activity and AngI and AngII levels that was associated with a decrease in systemic blood pressure without significant reflex tachycardia. The pharmacology of aliskiren is presented in more detail in Chapter 26.

Vasopressin Receptor Antagonists. Neurohumoral dysregulation in CHF extends beyond the renin– angiotensin–aldosterone axis to include abnormal arginine vasopressin (AVP) secretion, resulting in the perturbation of fluid balance, among various additional pathophysiogic effects. In response to 1) serum hypertonicity-induced activation of anterior pituitary osmoreceptors and 2) a perceived drop in blood pressure detected by baroreceptors in the carotid artery, aortic arch, and left atrium, AVP is secreted into the systemic circulation (Arai et al., 2007). The physiology and pharmacology of vasopressin are presented in Chapter 25.

The active form of AVP is a 9–amino acid peptide that interacts with three receptor subtypes: V_1a, V_1b, and V₂ (see Chapter 25). The AVP-V₂ receptor interaction on the basolateral membrane of the renal collecting ducts stimulates de novo synthesis of aquaporin-2 water channels that mediate free water reabsorption, thereby impairing diuresis and, ultimately, correcting plasma hypertonicity. Additional cell signaling pathways important in the pathophysiology of CHF include vasoconstriction, cell hypertrophy, and increased platelet aggregation mediated by activation of V_1a receptors in vascular smooth muscle cells and cardiac myocytes. In addition, AngII-mediated activation of centrally located AT₁ receptors is associated with increased AVP levels in CHF and may represent one mechanism by which the use of AT₁-receptor antagonists are effective in the clinical management of these patients.

β Adrenergic Receptor Antagonists

Sympathetic nervous system activation in CHF supports circulatory function by enhancing contractility (inotropy), augmenting ventricular relaxation and filling (lusitropy), and increasing heart rate (chronotropy). For many years, pharmacologic approaches to CHF treatment targeted drugs with sympathomimetic properties. This reflected the viewpoint that CHF is fundamentally a disorder of impaired stroke volume and cardiac output. For example, CHF symptom relief from short-term dobutamine and dopamine use in patients with ventricular dysfunction led to the belief that long-term sympathomimetic use would further improve clinical outcome. Under this model, the use of β receptor antagonists was believed to be counterproductive; however, the reverse appears to be the case. Long-term sympathomimetic use is associated with increased CHF mortality rates, whereas a survival benefit is associated with chronic administration of β receptor antagonists. Initially, clinical investigation of β receptor antagonists in the treatment of CHF encountered skepticism, but reports beginning in the early 1990s demonstrated that β antagonists (e.g., metoprolol) improve symptoms, exercise tolerance, and are measures of LV function over several months in idiopathic dilated cardiomyopathy patients with CHF (Waagstein et al., 1993; Gottlieb et al., 1998; Swedberg, 1993; Bristow, 2000). Serial echocardiographic measurements in CHF patients indicate that a decrease in systolic function occurs immediately after initiation of a β antagonist treatment, but this recovers and improves beyond baseline over the ensuing 2-4 months (Hall et al., 1995). This trend may be due to attenuation or prevention of the β receptor–mediated adverse effects of catecholamines on the myocardium (Eichhorn and Bristow, 1996).

Mechanism of Action. The mechanisms by which β receptor antagonists influence outcome in CHF patients are not fully delineated. By preventing myocardial ischemia without significantly influencing serum electrolytes, β receptor antagonists probably influence mortality, in part, by decreasing the frequency of unstable tachyarrhythmias to which CHF patients are particularly prone. In addition, these agents may influence survival by favorably affecting LV geometry, specifically by decreasing LV chamber size and increasing LV ejection fraction. Through inhibition of sustained sympathetic nervous system activation, these agents prevent or delay progression of myocardial contractile dysfunction by inhibiting maladaptive proliferative cell signaling in the myocardium, reducing catecholamine-induced cardiomyocyte toxicity, and decreasing myocyte apoptosis (Communal et al., 1998; Bisognano et al., 2000). β Receptor antagonists may also induce positive LV remodeling by decreasing oxidative stress in the myocardium (Sawyer and Colucci, 2000).

Metoprolol. Metoprolol (lopressor, toprol xl, others) is a β₁-selective receptor antagonist. The short-acting form of this drug has a drug elimination t_1/2 of ~6 hours, and therefore appropriate dosing is three to four times daily. Conversely, the extended-release formulation is sufficiently dosed once daily.

Carvedilol. Carvedilol (coreg, others) is a nonselective β receptor antagonist and α₁-selective antagonist that is FDA approved for the management of mild-to-severe CHF.

Clinical Use of β Adrenergic Receptor Antagonists in Heart Failure

Data from more than 15,000 patients with mild-to-moderate chronic CHF enrolled in various clinical trials have established that β receptor antagonists improve disease-associated symptoms, hospitalization, and mortality. Accordingly, β antagonists are recommended for use in patients with an LV ejection fraction <35% and NYHA class II or III symptoms in conjunction with an ACE inhibitor or AT₁ receptor antagonist, and diuretics as required to palliate symptoms. Interpretation of these and other clinical guidelines should, however, consider the following areas of uncertainty.

The role of β receptor antagonists in severe CHF or under circumstances of an acute clinical decompensation is not yet clear. Likewise, the utility of β blockade in patients with asymptomatic LV dysfunction has not been systematically evaluated. The marked heterogeneous pharmacologic characteristics (e.g., receptor selectivity, pharmacokinetics) of specific agents within this general drug class, as discussed in Chapter 12, play a key role in predicting the overall efficacy of a particular β receptor antagonist.

β receptor antagonist therapy is customarily initiated at very low doses, generally less than one-tenth of the final target dose, and titrated cautiously upward. Even when initiated properly, a tendency to retain fluid exists that may require diuretic dose adjustment. Insufficient evidence exists to support the unrestricted administration of β receptor antagonists in patients with severe (NYHA class IIIB and IV), new-onset, or acutely decompensated CHF.

Cardiac Glycosides

The English botanist, chemist, and physician Sir William Withering is credited with the first published observation in 1785 that digitalis purpurea, a derivative of the purple foxglove flower, could be used for the treatment of "cardiac dropsy," or congestive heart failure. The benefits of cardiac glycosides in CHF have been extensively studied (Eichhorn and Gheorghiade, 2002) and are generally attributed to:

• Inhibition of the plasma membrane Na⁺, K⁺-ATPase in myocytes

• A positive inotropic effect on the failing myocardium

• Suppression of rapid ventricular rate response in CHF-associated atrial fibrillation

• Regulation of downstream deleterious effects of sympathetic nervous system overactivation

Mechanism of the Positive Inotropic Effect. With each cardiac myocyte depolarization, Na⁺ and Ca²⁺ ions shift into the intracellular space (Figure 28–5). Ca²⁺ that enters the cell via the L-type Ca²⁺ channel during depolarization triggers the release of stored intracellular Ca²⁺ from the sarcoplasmic reticulum via the ryanodine receptor (RyR). This Ca²⁺-induced Ca²⁺ release increases the level of cytosolic Ca²⁺ available for interaction with myocyte contractile proteins, ultimately increasing myocardial contraction force. During myocyte repolarization and relaxation, cellular Ca²⁺ is re-sequestered by the sarcoplasmic reticular Ca²⁺-ATPase and is removed from the cell by the Na⁺Ca²⁺ exchanger and, to a much lesser extent, by the sarcolemmal Ca²⁺-ATPase.

Cardiac glycosides bind and inhibit the phosphorylated (α subunit of the sarcolemmal Na⁺,K⁺-ATPase and thereby decreasing Na⁺ extrusion and increasing cytosolic [Na⁺]. This decreases the transmembrane Na⁺ gradient that drives Na⁺−Ca²⁺ exchange during myocyte repolarization. As a consequence, less Ca²⁺ is removed from the cell and more Ca²⁺ is accumulated in the sarcoplasmic reticulum (SR) by SERCA2. This increase in releasable Ca²⁺ (from the SR) is the mechanism by which cardiac glycosides enhance myocardial contractility. Elevated extracellular K⁺ levels (i.e., hyperkalemia) cause dephosphorylation of the ATPase α subunit, altering the site of action of the most commonly used cardiac glycoside, digoxin, and thereby reducing the drug's binding and effect.

Pharmacokinetics. The elimination t_1/2 for digoxin is 36-48 hours in patients with normal or near-normal renal function, permitting once-daily dosing. Near steady-state blood levels are achieved ~7 days after initiation of maintenance therapy. Digoxin is excreted by the kidney, and increases in cardiac output or renal blood flow from vasodilator therapy or sympathomimetic agents may increase renal digoxin clearance, necessitating adjustment of daily maintenance doses. The volume of distribution and drug clearance rate are both decreased in elderly patients.

Clinical Use of Digoxin in Heart Failure. Data from contemporary clinical trials have re-characterized the utility of cardiac glycosides, once first-line agents, in CHF, especially in patients with normal sinus rhythm (as opposed to atrial fibrillation).

Digoxin Toxicity. The incidence and severity of digoxin toxicity have declined substantially in the past 2 decades as a consequence of alternative drugs available for the treatment of supraventricular arrhythmias in CHF, increased understanding of digoxin pharmacokinetics, improved serum digoxin level monitoring, and identification of important interactions between digoxin and other commonly co-administered drugs. Nevertheless, the recognition of digoxin toxicity remains an important consideration in the differential diagnosis of arrhythmias, and neurologic or gastrointestinal symptoms in patients receiving cardiac glycosides. An antidote, digoxin immune Fab (digibind, digifab), is available to treat toxicity.

β Adrenergic and Dopaminergic Agonists

In the setting of severely decompensated CHF from reduced cardiac output, the principal focus of initial therapy is to increase myocardial contractility. Dopamine and dobutamine are positive inotropic agents most often used to accomplish this. These drugs provide short-term circulatory support in advanced CHF via stimulation of cardiac myocyte dopamine (D₁) and β adrenergic receptors that stimulate the G_s-adenylyl cyclase-cyclic AMP–PKA pathway. The catalytic subunit of PKA phosphorylates a number of substrates that enhance Ca⁺²-dependent myocardial contraction and accelerate relaxation (Figure 28–5). Isoproterenol, epinephrine, and norepinephrine are useful in certain circumstances but have little role in routine CHF management. Indeed, inotropic agents that elevate cardiac cell cyclic AMP are consistently associated with increased risks of hospitalization and death, particularly in patients with NYHA class IV. At the cellular level, enhanced cyclic AMP levels have been associated with apoptosis (Brunton, 2005; Yan et al., 2007). The basic pharmacology of adrenergic agonists is discussed in Chapter 12.

Dopamine. Dopamine is an endogenous catecholamine with only limited utility in the treatment of most patients with cardiogenic circulatory failure. The pharmacologic and hemodynamic effects of dopamine are concentration dependent. Low doses (≤2 μg/kg lean body mass/min) induces cyclic AMP–dependent vascular smooth muscle vasodilation. In addition, activation of D₂ receptors on sympathetic nerves in the peripheral circulation at these concentrations also inhibits NE release and reduces α adrenergic stimulation of vascular smooth muscle, particularly in splanchnic and renal arterial beds. Therefore, low-dose dopamine infusion often is used to increase renal blood flow and thereby maintain an adequate glomerular filtration rate in hospitalized CHF patients with impaired renal function refractory to diuretics. Dopamine also exhibits a pro-diuretic effect directly on renal tubular epithelial cells that contributes to volume reduction.

At intermediate infusion rates (2-5 μg/kg/min), dopamine directly stimulates cardiac α receptors and vascular sympathetic neurons that enhance myocardial contractility and neural NE release. At higher infusion rates (5-15 μg/kg/min), α adrenergic receptor stimulation–mediated peripheral arterial and venous constriction occurs. This may be desirable in patients with critically reduced arterial pressure or in those with circulatory failure from severe vasodilation (e.g., sepsis, anaphylaxis). However, high-dose dopamine infusion has little role in the treatment of patients with primary cardiac contractile dysfunction; in this setting, increased vasoconstriction will lead to increased afterload and worsening of LV performance. Tachycardia, which is more pronounced with dopamine than with dobutamine, may actually provoke ischemia (and ischemia-induced malignant arrhythmias) in patients with coronary artery disease.

Dobutamine. Dobutamine is the β agonist of choice for the management of CHF patients with systolic dysfunction. In the formulation available for clinical use, dobutamine is a racemic mixture that stimulates both β₁ and β₂ receptor subtypes. In addition, the (−) enantiomer is an agonist for α adrenergic receptors, whereas the (+) enantiomer is a weak, partial agonist. At infusion rates that result in a positive inotropic effect in humans, the β₁ adrenergic effect in the myocardium predominates. In the vasculature, the α adrenergic agonist effect of the (−) enantiomer appears to be offset by the (+) enantiomer and vasodilating effects of β₂ receptor stimulation. Thus, the principal hemodynamic effect of dobutamine is an increase in stroke volume from positive inotropy, although β₂ receptor activation may cause a decrease in systemic vascular resistance and, therefore, mean arterial pressure. Despite increases in cardiac output, there is relatively little chronotropic effect.

Continuous dobutamine infusions are typically initiated at 2-3 μg/kg/min without a loading dose and uptitrated until the desired hemodynamic response is achieved. Pharmacologic tolerance may limit infusion efficacy beyond 4 days, and, therefore, addition or substitution with a class III PDE inhibitor may be necessary to maintain adequate circulatory support. The major side effects of dobutamine are tachycardia and supraventricular or ventricular arrhythmias, which may require a reduction in dosage. Recent β receptor antagonist use is a common cause of blunted clinical responsiveness to dobutamine.

Phosphodiesterase Inhibitors. The cyclic AMP–PDE inhibitors decrease cellular cyclic AMP degradation, resulting in elevated levels of cyclic AMP in cardiac and smooth muscle myocytes. The physiologic effects of this are positive myocardial inotropism and dilation of resistance and capacitance vessels. Collectively, therefore, PDE inhibition improves cardiac output through ionotropy and by decreasing preload and afterload (thus giving rise to the term inodilator). The clinical application of early-generation PDE inhibitors (e.g., theophylline, caffeine) is limited by low cardiovascular specificity and an unfavorable side-effect profile, whereas inamrinone, milrinone, and other more recently developed PDE inhibitors are preferred.

Chronic Positive Inotropic Therapy

Several orally administered agents with combined inotropic and vasodilator properties are available for clinical use. Although improvements in CHF symptoms, functional status, and hemodynamic profile have been reported, the effect of long-term therapy on mortality is disappointing. In fact, the dopaminergic agonist ibopamine; PDE inhibitors milrinone, inamrinone, and vesnarinone; and pimobendan are associated with increased mortality (Hampton et al., 1997; Packer et al., 1991; Cohn et al., 1998). At present, digoxin remains the only oral inotropic agent available for CHF patient use.

Diastolic Heart Failure

Data from population studies suggest that up to 40% of CHF patients have preserved LV systolic function. The pathogenesis of diastolic CHF includes structural and functional abnormalities of the ventricle(s) that are associated with impaired ventricular relaxation and LV distensibility. These abnormalities are reflected in the LV pressure–volume relationship during diastole, which is shifted upward and to the left relative to normal subjects (Figure 28–6). Consonant with the definition of CHF outlined earlier in this chapter, the diagnosis of diastolic CHF is made when the LV is unable to maintain adequate cardiac output without filling at an abnormally elevated end-diastolic filling pressure.

In patients with primary diastolic dysfunction, the myocardial abnormality that accounts for abnormal filling is intrinsic to the myocardium; e.g., by infiltrative disorders including cardiac amyloidosis, hemochromatosis, sarcoidosis, and rarer conditions such as endomyocardial fibrosis and Fabry's disease. Although not a disease of myocardium infiltration, clinically evident CHF may occur despite intact LV systolic function in familial hypertrophic cardiomyopathy.

Secondary diastolic dysfunction occurs as a consequence of excessive preload (e.g., renal failure), excessive afterload (e.g., systemic hypertension), or changes in LV geometry that occur in response to chronically abnormal loading conditions. Diastolic CHF also is observed in patients with long-standing epicardial coronary artery or pericardial disease. The prevalence of secondary diastolic dysfunction is higher in women and with advanced age. Reported annual mortality rates for diastolic CHF are 5-8%, although this range likely represents an underestimation (Jones et al., 2004).

Patients with diastolic CHF are typically dependent on preload to maintain adequate cardiac output. Although hypervolemic patients generally benefit from careful intravascular volume reduction, this should be accomplished gradually and treatment goals reassessed frequently. Maintaining synchronous atrial contraction (or at least ventricular rate response control) helps to maintain adequate LV filling during the latter phase of diastole and is therefore a paramount goal in the management of CHF from diastolic dysfunction. Evaluation and treatment of predisposing conditions to impaired diastolic function, such as myocardial ischemia and poorly controlled systemic hypertension, are fundamental to the overall pharmacotherapeutic strategy of this complex form of CHF.

Future Therapies: Targeting Vascular Dysfunction in Congestive Heart Failure from Systolic Dysfunction

Vascular dysfunction is an established component of the CHF syndrome and has evolved into a novel pharmacotherapeutic target for the clinical management of patients with this disease (Varin et al., 2000) (Figure 28–7). Contemporary scientific observations suggest that the blood vessel is a dynamic structure integral to normal myocardial function. This represents a paradigm shift away from the traditional perspective that blood vessels are conduit "tubes" necessary only for blood transport. Elevated levels of oxidant, nitrosative, and other forms of inflammatory stress observed in patients with CHF may impair vascular reactivity by disruption of normal vasodilatory cell signaling pathways (Erwin et al., 2005; Doehner et al., 2001). The precise mechanism by which impaired vascular reactivity is aligned with the progressive natural history of CHF is unresolved; when present, however, vascular dysfunction is associated with decreased exercise tolerance and a poorer clinical outcome. For example, hyperaldosteronism due to overactivation of the renin–angiotensin–aldosterone axis in the setting of LV dysfunction adversely affects both endothelium-dependent and endothelium-independent vascular reactivity (Leopold et al., 2007, Maron et al., 2009) enfothelium XO-I zenthine oxidase inhibitor vascular reactivity. This process is in part mediated by increased levels of reactive oxygen species, decreased endogenous levels of antioxidant enzymes, and decreased levels of bioavailable NO (Leopold et al., 2007; Farquharson and Struthers, 2000). As discussed previously, the undesirable effects of hyperaldosteronism on vascular dysfunction are attenuated clinically by aldosterone receptor blockade, resulting in significantly decreased CHF-associated morbidity and mortality (Pitt et al., 1999).

Xanthine Oxidase and Vascular Dysfunction. Xanthine oxidase (XO) is necessary for normal purine metabolism and catalyzes the oxidation of hypoxanthine to xanthine and xanthine to uric acid in a reaction that generates superoxide. Elevated levels of uric acid are associated with clinically evident CHF (Hare et al., 2008). For example, epidemiologic data support a positive, graded association between impaired exercise capacity and circulating uric acid levels (Doehner et al., 2001). Although the myocardium is rich in XO, vascular endothelial cells also contain high concentrations of XO, an observation that ultimately lead to the hypothesis that increased XO-generated superoxide impairs vascular reactivity in CHF patients (Hare et al., 2008).

Statins and Vascular Dysfunction. HMG-CoA (3-hydroxy-3-methyl-glutaryl–coenzyme A) reductase catalyzes the formation of L-mevalonic acid, a key biochemical precursor in the cholesterol synthesis pathway (Goldstein and Brown, 1990). Current evidence suggests a role of crosstalk between mevalonate metabolism and cell signaling pathways involved in inflammation and oxidant stress. In this way, HMG-CoA reductase inhibitors, or "statins", may exert beneficial cardiovascular effects beyond their original intent of low-density lipoprotein reduction; specifically, statins are associated with positive LV remodeling, increased arteriolar blood flow, and decreased circulating platelet aggregation (Liao et al., 2005).

Intermediate byproducts of mevalonate metabolism (i.e., isopenylated proteins that upregulate activation of Rho, RAS, and other G proteins) are linked to impaired vascular function by increasing levels of oxidant stress and decreasing bioavailable NO levels (Hernandez-Perera et al., 1998). Statins inhibit these intermediary pathways and appear to restore endothelium-dependent (Feron et al., 2001) and endothelium-independent vascular function (Drexler et al., 1993). The pharmacology of the statins and other cholesterol-lowering agents is presented in Chapter 31.

Direct Activators of Soluble Guanylyl Cyclase. Soluble guanylyl cyclase (sGC) is an enzyme that catalyzes the conversion of guanosine triphosphate to cyclic GMP, a second messenger necessary for normal vascular smooth muscle cell relaxation (Koesling, 1999). Under physiologic conditions, NO is the primary biologically active stimulator of sGC. Elevated levels of oxidant stress deactivate sGC through various molecular mechanisms. For example, aldosterone levels comparable to those observed in patients with decompensated CHF are associated with increased oxidant stress that converts sGC to an NO-insensitive state (Maron et al., 2009), thereby disrupting vasodilatory signaling necessary for normal vascular function.

Organic nitrates, which promote sGC activation by increasing bioavailable NO levels, are subject to pharmacologic tolerance that complicates long-term drug use, dosing, and administration frequency (see the organic nitrates section in Chapter 27). Although the precise mechanism to explain this phenomenon is unknown, it is likely mediated in part by elevated levels of oxidant stress that convert sGC to an NO-insensitive state. BAY compounds (e.g., BAY 58-2667 [cinaciguat]) activate sGC by an NO-independent mechanism, thereby promoting normal sGC function despite conditions of oxidant stress (Stasch et al., 2006). Data from preclinical CHF studies in animals have validated these beneficial molecular effects. In healthy humans, BAY compound administration has not been associated with severe side effects, but hypotension and headache have been reported. Likewise in humans, circulating plasma drug concentrations are decreased to clinically insignificant levels ≤30 minutes after infusion termination (Frey et al., 2008).

Congestive heart failure is a chronic and usually progressive illness instigated by a primary myocardial insult that impairs myocardial structure and function. Treatment (Figure 28–8) ideally begins with prevention of cardiac dysfunction via the identification and remediation of risk factors that predispose individuals to the development of structural heart disease. For example, because coronary artery disease is the most common etiology of LV systolic dysfunction, public health tactics to address strategies for prevention of MI through lipid lowering, blood pressure control, and smoking cessation are critical for effective CHF primary prevention.

In asymptomatic patients with LV dysfunction or impaired diastolic dysfunction, neurohormonal compensatory mechanisms that support cardiovascular function are activated, but, paradoxically, further disease progression. Indeed, treatment in this clinical stage is directed at attenuation of sustained neurohormonal activation, with individualized regimens of β adrenergic receptor blockers and ACE inhibitors or AT₁ receptor antagonists.

In most cases, there is unwanted progression to clinically evident CHF, indicating the development of significantly abnormal cardiovascular hemodynamics. Treatment goals in these patients include symptom relief and prevention of disease progression. This may be achieved with diuretics, load-reducing agents, and renin–angiotensin–aldosterone axis antagonists.

Symptomatic patients with hemodynamic decompensation require hospitalization, because oral diuretic and vasodilator therapy alone often are inadequate to reestablish euvolemia and adequate peripheral perfusion. Parenteral vasodilator and inotropic agents are administered parenterally to restore forward cardiac output and to reverse Na⁺ and water retention. Effective treatment of these patients often is complicated by concurrent renal insufficiency and hypotension, despite evidence of persistently elevated intracardiac filling pressures. In this setting, selection of parenteral agents guided by data obtained from an indwelling pulmonary artery catheter may prove useful.

Consider as an example the clinical scenario in which a decompensated CHF patient presents with normal systemic vascular resistance. In this case, afterload reduction may be contraindicated in the short term and treatment with a parenteral agent such as dobutamine may be preferable. The risk attendant to treatment with sympathomimetic drugs is related to the increase in myocardial O₂ consumption that may occur; this is of particular concern in patients with left-heart failure that occurs as a direct consequence of myocardial ischemia. This clinical quandary has become less common in the era of aggressive myocardial revascularization; when it is encountered, coadministration of dobutamine with parenteral nitroglycerin may be necessary.

In CHF patients who are persistently symptomatic or hemodynamically unstable, referral to a specialized tertiary center with expertise in the evaluation and management of end-stage CHF should be considered. Under these circumstances, resource-intensive treatment, including cardiac resynchronization therapy, mechanical assist devices, high-risk surgical interventions, and cardiac transplantation, may be considered.

Despite the numerous pharmacotherapeutic advances outlined in this chapter, a poor prognosis still remains for patients suffering from severe or end-stage CHF. For these patients, the empathetic initiation of end-of-life discussions may be useful in order to establish appropriate treatment expectations compatible with effective and patient-centered medical care.