CHAPTER 71: EVALUATION AND MANAGEMENT OF ACUTE HEART FAILURE

Prevalence

Currently, the prevalence of heart failure (HF) in the United States is approaching 6 million, a number that will continue to grow as the population ages.¹ Despite advances in pharmacologic and device therapies, HF results in significant mortality and morbidity, including frequent emergency department visits and hospitalizations for decompensation. Acute heart failure (AHF) accounts for an estimated 700,000 emergency department visits² and over 1 million hospitalizations annually in the United States.¹ It is the leading cause of hospitalization in patients over 65 years of age, a population expected to grow considerably over the next 20 years.¹ Highlighting the prevalence of comorbid conditions in the HF population, an additional 3 million hospitalizations occur annually in which HF is listed as a secondary diagnosis.³ In a large, community-based study, the incidence of AHF was as high as 11.6 cases per 1000 patients per year in those over the age of 55 years.⁴ The economic impact of the HF epidemic is profound, and costs attributable to inpatient care are estimated to exceed 30 billion dollars annually.⁵

Data from large registries reveals that the typical AHF patient is > 70 years old, white (~80%), and equally likely to be female (~50%) compared to male.^6,7 Although most AHF patients are white, in the community setting, black males have the highest incidence of decompensation (15.7 per 1000 people per year), followed by black women (13.3 per 1000 people per year), white men (12.3 per 1000 people per year), and white women (9.9 per 1000 people per year).⁴

Although historically attention has focused on patients with systolic dysfunction (HF with reduced ejection fraction [HFrEF]), it is now clear that over 50% of AHF admissions occur in those with preserved ejection fraction (HF with preserved ejection fraction [HFpEF]).⁸ Additionally, the proportion of hospitalizations in those with HFpEF is increasing considerably compared to those with HFrEF, likely a reflection of the aging population.⁹ Several differences exist between these two groups, with HFpEF patients being more likely to be older and female, more likely to have less clinical coronary artery disease (CAD), and more likely to have atrial fibrillation, chronic kidney disease, higher blood pressure, and chronic pulmonary disease compared to their HFrEF counterparts.⁹ Importantly, HFpEF patients with AHF have greater hospital lengths of stay and are more likely to be discharged to a skilled nursing facility.⁹

Importance of Comorbidities

Comorbidities are common in patients presenting with AHF, including systemic hypertension (> 70%),¹⁰ CAD (57%),² diabetes mellitus (40%),⁷ renal dysfunction (30%-67%),¹¹ anemia (50%-70%),¹² atrial fibrillation (31%),⁷ and chronic obstructive pulmonary disease (COPD; 30%),¹⁰ among others. These comorbidities are associated with higher inpatient mortality, regardless of the presence of preserved or reduced ejection fraction.^13,14 They may also confound the diagnosis of AHF (eg, COPD), reduce the effectiveness of treatments targeting decongestion (eg, renal dysfunction), and exacerbate an otherwise stable outpatient (eg, atrial fibrillation, uncontrolled hypertension). Thus, treatment of AHF almost universally involves management of comorbidities. An analysis of patients > 65 years old presenting with AHF revealed that 40% have greater than five comorbid conditions, leading to greater resource utilization and lengths of stay.¹⁴

Outcomes

AHF is a harbinger of poor outcomes and is associated with significant mortality after discharge. Recent trends suggest in-hospital mortality ranges from 2% to 5% and are similar when comparing HFpEF to HFrEF.^9,15 Postdischarge mortality for these two groups approaches 10% at 90-day follow-up,¹⁶ whereas approximately 30% of patients die within 1 year of hospitalization (Fig. 71–1).^15,17 Despite concerted efforts to reduce rehospitalizations for AHF, nearly 25% of patients are readmitted within 30 days of discharge, and 50% are readmitted by 6 months.^18,19 Each subsequent hospitalization following the index stay is associated with increasing risk of death (Fig. 71–2).

Although there is a tendency to treat patients with AHF as a homogenous group, it is clear that the syndrome of decompensated HF is complex and the “one size fits all” adage does not apply. Indeed, patients presenting with AHF are a heterogeneous group with a variety of underlying cardiac substrates and abnormalities. Additionally, the severity of AHF exists on a spectrum, ranging from mild fluid overload to overt cardiogenic shock, and therefore, no singular pathophysiologic model can account for the diversity of clinical presentations. Nevertheless, several key factors play varying roles in the initiation and propagation of AHF, including but not limited to: (1) acute or subacute worsening of cardiac dysfunction, either systolic or diastolic; (2) neurohormonal activation; (3) the cardiorenal syndrome resulting from the complex interplay between the kidney and the heart; (4) impaired vascular/endothelial function; (5) inflammation and oxidative stress; and (6) hemodynamic factors. Ultimately, AHF is the end product of further comprise of an abnormal cardiac substrate, a precipitant that triggers instability, and a self-sustaining cycle of maladaptive mechanisms (Fig. 71–3).

Abnormal Substrate: Role of Cardiac Dysfunction

HF is defined as a dysfunction of the heart that impairs its ability to maintain the metabolic needs of the body in the absence of elevated filling pressures. This dysfunction may manifest as impaired contractility (systolic dysfunction), impaired relaxation or decreased ventricular compliance (diastolic dysfunction), or more commonly, both. AHF in the setting of contractile dysfunction, often referred to clinically as acute systolic HF, may result from an abrupt insult (eg, myocardial infarction, myocarditis) or, more commonly, a worsening of a chronic, progressive process. The hallmark features of acute systolic HF include elevated left ventricular (LV) filling pressures, manifesting as congestion and dyspnea, and either subtle or overt impairments in systemic perfusion pressure (“arterial underfilling”), which may manifest as hypoperfusion or end-organ damage in extreme cases (eg, cardiogenic shock). When occurring abruptly as in the case of a myocardial infarction, cardiac output (CO) may be significantly reduced even with relatively small infarcts; however, when occurring progressively, systolic dysfunction typically does not lead to overt reductions in CO until the late stages, related to the adaptive mechanisms at play to compensate for impaired ventricular function (ie, cardiac remodeling). Notably, even subtle abnormalities of systemic perfusion may result in upregulation of evolutionarily “protective” mechanisms to augment circulating blood volume and maintain mean arterial pressure, namely the renin-angiotensin-aldosterone system (RAAS), the sympathetic nervous system (SNS), and pituitary vasopressin release. Although this so-called “neurohormonal activation” may serve to acutely augment perfusion pressures, it ultimately leads to volume overload, myocardial damage and fibrosis, and adverse ventricular remodeling, further amplifying the HF syndrome.

AHF in the absence of overt systolic dysfunction, termed acute diastolic HF, is largely a manifestation of the inability of the LV to effectively fill in diastole, be it passively or actively. This may occur abruptly (eg, acute onset atrial fibrillation with rapid ventricular response, accelerated hypertension) or progressively over time (eg, worsening hypertensive heart disease, progressive aortic stenosis). As diastolic dysfunction progresses (be it acute or chronic), left atrial pressure become increasingly elevated to facilitate ventricular filling. This escalating pressure is reflected upstream to the pulmonary capillary circulation, leading to pulmonary congestion and dyspnea. Although ejection fraction is preserved in acute diastolic HF by definition, the systolic function as measured by effective stroke volume at rest or exercise is often abnormal in many of these patients. Additionally, these patients often have evidence of increased ventricular systolic and arterial stiffness, which may manifest as lower stroke volume and contractile reserve.²⁰ The end product of these abnormalities is similar to patients with acute systolic HF: activation of RAAS, SNS, and vasopressin leading to worsening fluid overload and the ongoing propagation of the HF syndrome.

Neurohormonal Activation in Acute Heart Failure

The primary objective of the circulatory system is to maintain perfusion to vital organs by sustaining arterial pressure and circulating blood volume. Evolutionarily, a number of mechanisms that maintain organ perfusion have been identified, including the RAAS axis, the SNS, and the pituitary release of vasopressin (antidiuretic hormone [ADH]). Although acute upregulation of these systems is effective at restoring circulating blood volume and perfusion pressures in the short term, chronic activation leads to sodium and fluid overload as well as cardiac fibrosis and remodeling.²¹

When perfusion pressures are reduced, whether as a result of acute blood loss or HF, the juxtaglomerular cells of the nephron release renin into the circulation. Renin converts angiotensinogen to angiotensin I, which is then converted to angiotensin II by angiotensin-converting enzyme (ACE). Angiotensin II has multiple downstream effects, including: (1) increased systemic vascular resistance through vasoconstriction; (2) volume expansion by promoting sodium reabsorption in the proximal tubule; (3) stimulation of adrenal aldosterone release; and (4) activation of the SNS. Aldosterone release from the adrenal glands leads to additional sodium resorption in the distal tubule and collecting duct, volume expansion, and ultimately myocardial fibrosis. Thus, RAAS activation plays a fundamental role in AHF through fluid retention and vasoconstriction while also contributing to chronic ventricular remodeling.

Reductions in perfusion pressure also lead to activation of the SNS, primarily through the suppression of cardioinhibitory baroreceptor reflexes and the augmentation of excitatory reflexes.²² SNS activation results in increased CO by increasing myocardial contractility and heart rate, while also increasing blood pressure through norepinephrine-mediated vasoconstriction. Additionally, heightened sympathetic activity leads to renin release from the kidney and the propagation of the RAAS axis, further exacerbating AHF. Much like the RAAS axis, the acute effects of SNS activation favorably support organ perfusion. However, chronic catecholamine exposure leads to myocyte death, myocardial fibrosis, and adverse cardiac remodeling.

Finally, ADH is released from the posterior pituitary in response to both osmotic and nonosmotic mechanisms. Angiotensin II triggers ADH release, as does a drop in perfusion pressure. ADH augments free water absorption in the kidney through aquaporin upregulation, stimulates thirst centers in the brain that trigger fluid intake, and increases vascular resistance through vasoconstriction. Ultimately this leads to free water overload, altered hemodynamics, and worsening AHF.

Renal Dysfunction

Renal dysfunction is common in AHF, affecting between 30% and 67% of patients in large registries.¹¹ Additionally, as many as 20% to 40% of patients experience worsening renal function (WRF) while being treated for AHF, a finding associated with worse prognosis.^11,23,24,25 Although many consider the kidneys the “barometer” of the heart in AHF, our understanding of the complex interactions between these two organs remains poor. The pathophysiology of renal dysfunction in HF likely incorporates hemodynamic factors, activation of the RAAS and SNS, inflammation and oxidative stress, and the iatrogenic effects of pharmacotherapy targeting decongestion, among other factors (Fig. 71–4).

While historically WRF in AHF was considered a consequence of reduced CO (“underperfusion”), it has become increasingly apparent that right-sided congestion (eg, right atrial pressure) plays an important role. Multiple studies have failed to show a correlation between CO and WRF in AHF,^26,27,28 with the exception of extreme reductions in output.²⁹ In contrast, several studies show a correlation between elevated right atrial pressure and intra-abdominal pressure with WRF.^26,27,28,30

Neurohormonal activation plays a critical role in WRF in the setting of decompensation. The downstream effects of RAAS and SNS activation include angiotensin II and catecholamine-induced renal efferent arteriolar constriction, which increases glomerular filtration at the expense of renal blood flow. Additionally, stimulation of aldosterone secretion increases sodium resorption and exacerbates volume overload and congestion, which in turn promotes WRF. Chronic RAAS and SNS activation may lead to renal interstitial fibrosis and inflammation, resulting in chronic kidney disease.³¹

A number of pharmacologic treatments of AHF may result in WRF. ACE inhibitors have beneficial hemodynamics effects in AHF (reductions in preload and afterload), but are associated with a reduction in glomerular filtration rate as high as 30%.³² Thus, they must be used with caution in those who already have considerable renal dysfunction, and further studies are needed to clarify the role of these agents specifically in AHF. Loop diuretics may also result in WRF. By increasing the delivery of sodium to the distal tubule, loop diuretics trigger adenosine release from the juxtaglomerular cells. The downstream effects of increased intrarenal adenosine include augmented sodium reabsorption in the proximal tubule (and therefore worsening congestion) and a reduction in glomerular filtration rate mediated by afferent arteriolar constriction. Unfortunately, treatments targeting adenosine receptors failed to prevent WRF in AHF in clinical trials.³³

Regardless of the precise mechanisms at play, the kidneys play a crucial role in the syndrome of AHF, primarily through retention of sodium, leading to volume overload and congestion. This congestion can then cause WRF, triggering a vicious cycle that leads to further cardiac decompensation. Additionally, renal dysfunction leads to less utilization of standard HF pharmacotherapy (eg, ACE inhibitors, mineralocorticoid receptor antagonists [MRAs]), higher diuretic requirements to achieve effective decongestion, and a higher risk for adverse events.¹¹ However, it should be noted that WRF in the face of aggressive diuresis and hemoconcentration in a patient who is clinically improving may represent simply a delayed equilibration of fluid between the extra- and intravascular compartments, and may be associated with improved and not worse outcomes.^34,35 Such transient WRF seems to result in better outcomes compared to those with persistent WRF during treatment for AHF.³⁶

Vascular and Endothelial Dysfunction

The role of the systemic vasculature in the pathophysiology of AHF has been known since the early 1960s, when studies showed elevations in afterload impair cardiac performance.^37,38 Many subsequent hemodynamic studies of both intravenous and oral vasodilator agents showed significant improvements in CO and filling pressures in patients with AHF treated with these drugs.³⁷ Increased afterload, or systemic vascular resistance (SVR), is thought to result from endothelial dysfunction, common in both chronic and acute HF syndromes. Endothelial dysfunction, a by-product of an underlying nitroso-redox imbalance in HF,^39,40 leads to systemic vasoconstriction and elevated SVR. These mechanisms are amplified in AHF, resulting in greater cardiac work and higher LV filling pressures. Thus, vasodilator agents aimed at reducing afterload are prescribed commonly in the management of AHF.

Inflammation and Oxidative Stress

Inflammation plays a key role in the pathophysiology of AHF and also has important prognostic significance. Several inflammatory cytokines are upregulated in AHF, including tumor necrosis factor (eg, TNF-α), interleukins (eg, IL-1, IL-6), and acute-phase reactant C-reactive protein (CRP). Many of these factors directly suppress cardiac contractility and promote myocyte death, thereby worsening AHF. Higher levels also correlate with increased mortality.⁴¹ Inflammation is associated with oxidative stress, manifest as an imbalance in nitric oxide (NO) and reactive oxygen species (ROS). This nitroso-redox imbalance leads to a variety of adverse effects that worsen the pathophysiology of AHF, including impairments in both ventricular and vascular function. Thus, there is a theoretical role for pharmacotherapy specifically targeting reductions in ROS (eg, hydralazine) and augmentation of NO (eg, isosorbide dinitrate) (Fig. 71–5).⁴²

Hemodynamics

The hemodynamics in AHF are highly variable and depend on the severity of the underlying substrate and the degree of decompensation. However, elevated filling pressures are the hallmark of hemodynamic aberrations in AHF. Increased LV end-diastolic pressure (LVEDP) results in elevated left atrial (LA) pressure, resulting in pulmonary congestion and dyspnea. Additionally, elevated right atrial pressure is common in AHF, accounting for lower extremity edema, ascites, liver congestion, and early satiety. Not uncommonly, secondary pulmonary hypertension may be present, initially reflecting passive hydrostatic forces from the left atrium to the pulmonary capillary bed. Over time, pulmonary vasoconstriction and adverse pulmonary arteriolar remodeling may occur, leading to pulmonary hypertension out of proportion to the LA pressure⁴³ and markedly elevated pulmonary vascular resistance.

In most cases of AHF, CO is in the normal range or only mildly reduced at rest. In the case of an acute process, such as a myocardial infarction or fulminant myocarditis, CO may drop dramatically, leading to a low CO syndrome or shock characterized by end-organ underperfusion requiring inotropic or mechanical circulatory support. When AHF develops from a progressive exacerbation of chronic LV dysfunction, CO is typically not severely reduced in the absence of very late-stage or advanced disease. SVR may be normal or significantly elevated in AHF. In cases of systolic dysfunction and elevated SVR, vasodilator therapy may reduce SVR and lead to improvements in stroke volume and LV filling pressures. This may not be true in patients with decompensated HFpEF. Recent physiologic studies suggest that aggressive afterload reduction in the setting of HFpEF may actually reduce stroke volume.⁴⁴ This drop in stroke volume likely results from a reduction in LV end-diastolic volume (preload), although the mechanisms at play are not entirely clear. Therefore, afterload reduction should be used with caution in this setting.

Although much progress has been made in the treatment of chronic HF, little has changed in the management of AHF in the past 40 years. As a result, mortality for patients with AHF remains unacceptably high.^9,15 However, not all patients face the same risk. A number of clinical variables may identify subgroups at risk for worse outcomes, and several risk scores have been developed to better facilitate stratification of patients with AHF (Table 71–1).^{45,46,47,48,49} Additionally, several novel biomarkers have emerged that have shown considerable promise in the risk stratification of patients presenting with AHF or being discharged following a hospitalization for AHF. Biomarkers will be more comprehensively discussed later in this chapter.

The Acute Decompensated Heart Failure National Registry (ADHERE) “risk tree” was derived from an analysis of 39 admission variables in over 60,000 AHF patients.⁴⁵ Investigators found that the best predictors of inpatient mortality were markers of renal dysfunction (elevated blood urea nitrogen and creatinine) and lower blood pressure. A risk tree was developed incorporating these variables, enabling clinicians to estimate inpatient mortality ranging from low risk (~2%) to intermediate risk (~6%-13%) to high risk (~20%) (Fig. 71–6).⁴⁵

The Enhanced Feedback for Effective Cardiac Treatment (EFFECT) risk index uses a number of demographic, laboratory, and clinical variables at the time of admission to determine death risk at 30 days and 1 year of follow-up.⁴⁶ Developed from a retrospective study of over 4000 community-based AHF patients in Ontario, Canada, the EFFECT risk index can be used to categorize patients into low risk (mortality 0.4% at 30 days and 7.8% at 1 year) and high risk for death during follow-up (mortality 59% at 30 days and 79% at 1 year). The variables incorporated into this risk index may be found in Table 71–1.⁴⁶

Finally, the Get With the Guidelines–Heart Failure (GWTG-HF) risk score was derived from a cohort of over 39,000 patients admitted with AHF between 2007 and 2009.⁴⁹ This score incorporates a number of admission variables associated with risk of death in a multivariable analysis: age, systolic blood pressure, blood urea nitrogen, heart rate, sodium, COPD, and nonblack race. This model has been validated for use across a wide spectrum of AHF patients and is effective in both HFpEF and HFrEF patients.⁴⁹ The above scores represent just a few of the many AHF risk scores available for clinical use. Additional information on scoring models is shown in Table 71–1.^{45,46,47,48,49}

Therapeutic options for chronic HFrEF are well delineated and are founded in strong clinical trial evidence. In contrast, therapy for AHF has been met with largely negative or neutral clinical trials and less established treatment protocols. As new treatment options for AHF emerge, there is a critical need to identify patients appropriate for various therapies. Careful consideration should be paid to the need for inpatient management.⁵⁰ A general approach to the treatment of AHF is addressed in the American College of Cardiology (ACC)/American Heart Association (AHA)⁵¹ guidelines, which include recommendations for the following: immediate determination of volume and perfusion status; addressing precipitating factors and comorbidities; recognition of acuity versus chronicity of the disease process; interpretation of contributing biomarker data; initiation or cautious uptitration of loop diuretic therapy without time delay; and initiation of intravenous inotropic or vasopressor drugs in those patients with systemic hypotension as indicated by acute kidney injury or change in mental status and severe elevation in cardiac filling pressures. Treatment of AHF centers on maintenance or restoration of systemic perfusion and preservation of end-organ performance and relieving congestion, while consideration of more definitive or advanced therapies takes place. In most patients admitted with AHF and HFrEF, chronic oral maintenance therapies, such as β-blocker (BB), ACE inhibitor (ACEI), or angiotensin receptor blocker (ARB), can be continued without interruption, assuming relative hemodynamic stability. In patients with HFrEF who are not already on neurohormonal antagonist medications, these agents should be initiated once hemodynamic stability and euvolemia have been achieved, because these therapies are known to improve outcomes⁵¹ (Table 71–2). Hospitalization provides an opportunity for timely and comprehensive assessment.⁵²

Classification

There is significant variability in patients who present with AHF, making appropriate classification paramount to appropriate triage and therapeutic targets. Although the majority of patients present with congestion, there is a significant variability in underlying cardiac and noncardiac comorbidities and variation in clinical presentation in AHF. There are many published classification systems for AHF (Table 71–3). The European Society of Cardiology (ESC) guidelines propose six clinical profiles, including: (1) worsening or decompensated chronic HF, (2) pulmonary edema, (3) hypertensive HF, (4) cardiogenic shock, (5) isolated right HF, and (6) acute coronary syndrome and HF.⁵³ The ACC/AHA describes three groups, including: (1) patients with volume overload (generally with pulmonary and/or systemic congestion and often hypertension), (2) those with hypotension and severe reduction in CO, and (3) those with combination cardiogenic shock and congestion.⁵⁴

Pang et al⁵⁵ describe a classification framework combining both ACC/AHA and ESC guidelines, highlighting the need for optimization of management of known targets and the importance of recognizing “responders” to therapy.⁵⁵ Gheorghiade and Braunwald⁵⁶ propose a six-axis model in which severity informs immediate versus deferred therapy and inpatient management, using readily available intake parameters. This model takes into account de novo versus worsening chronic HF, blood pressure (low, normal, or high), heart rate and rhythm, severity of dyspnea, comorbidities, and precipitants.⁵⁶ Felker et al^57B highlighted a need for classification with respect to clinical trial design, proposing a clinically useful classification framework in this context. The authors recommend separation into clinical profiles of AHF based on patients with AHF and hypertension, hypotension, volume overload, and other precipitating factors.^57B

Immediate Assessment

Signs, Symptoms, Physical Examination, and Hemodynamic Profiling

Dyspnea, or breathlessness, is the most common symptom reported by patients with AHF regardless of severity, followed by fatigue. In Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients With Heart Failure (OPTIMIZE-HF),⁵⁸ 64% of enrolled patients had rales on exam, likely reflecting increased pulmonary capillary wedge pressure (PCWP). Dyspnea can be assessed by the provocative dyspnea assessment, which uses varying positions and maneuvers to assess level of dyspnea under stress, with and without oxygen.⁵⁹ Pang et al⁶⁰ proposed a dyspnea severity score, based on provocative dyspnea assessment performance using a 5-point Likert scale to standardize degree of dyspnea in clinical trials. Signs and symptoms in AHF vary based on clinical phenotype and inform treatment decisions.⁵⁷ For example, patients in AHF who present with hypertension may present with primarily pulmonary congestion with or without systemic congestion, with acute pulmonary edema in the most severe cases. AHF patients with worsening chronic HF and volume overload are likely to present with peripheral edema and less likely pulmonary or radiographic congestion. Patients with AHF and hypotension suggestive of a low CO state may have symptoms of poor end-organ perfusion, such as confusion, lethargy, and abdominal discomfort. A small subset of these patients present in cardiogenic shock.⁵⁷ Those with isolated right HF are likely to present with absence of pulmonary congestion with signs of low LV filling pressures with or without hepatomegaly.⁵³

The physical exam in AHF must focus on vital signs (tachycardia, hyper- vs hypotension, narrow pulse pressure), heart and lung exam, neck veins, and assessment of volume overload in general.⁵¹ The presence of jugular venous distention suggests elevated right-sided filling pressures, which in 80% of cases suggests elevated left-sided filling pressures.⁶¹ In AHF, proportional pulse pressure ([systolic – diastolic blood pressure]/systolic blood pressure) < 25% is associated with low cardiac index (< 2.2 L/min/m²) with high sensitivity and specificity, and can therefore be an easy clinical marker.⁶² Cardiac exam may reveal a third heart sound, suggestive of LV dilatation, or fourth heart sound, which may indicate poor LV compliance. The Valsalva maneuver may be used to indicate the presence of altered arterial pressures, suggesting LV systolic dysfunction,⁶³ but may be more challenging to perform in the acute setting. New or worsened murmur may indicate worsened ventricular dilatation or new valvular disease. Lung exam can reveal rales or wheezing. Abdominal exam may reveal hepatomegaly signifying passive congestion, or ascites. The presence of a hepatojugular reflux can indicate right ventricular dysfunction. Extremity exam can provide insight on both congestion (presence of lower extremity and dependent edema) and perfusion (poor capillary refill, cool temperature) (Table 71–4).⁵⁹

Hemodynamic Profiling

The conventional approach to therapy in AHF begins with bedside determination of the hemodynamic profile. Nohria et al⁶⁴ described four patient profiles based on volume and perfusion status (Fig. 71–7). This simple classification has proven to be very useful in clinical practice.

• Profile A: Patients without evidence of congestion with adequate perfusion (“dry-warm”)

• Profile B: Patients with congestion but adequate perfusion (“wet-warm”)

• Profile C: Patients with congestion and hypoperfusion (“wet-cold”)

• Profile L: Patients without congestion, with hypoperfusion (“dry-cold”)⁶⁴

Survival analysis revealed that clinical profiles predicted outcomes in HF, with profiles B and C portending increased risk of death or urgent transplantation, adding prognostic value even when limited to patients with New York Heart Association (NYHA) class III and IV symptoms. The presence of congestion was determined by orthopnea and/or physical exam evidence of jugular venous distention, pulmonary rales, hepatojugular reflux, ascites, peripheral edema, leftward radiation of the pulmonic heart sound, or a square wave blood pressure response to the Valsalva maneuver. Perfusion status was determined by the presence of narrow pulse pressure ([systolic – diastolic]/systolic blood pressure < 25%), pulsus alternans, symptomatic hypotension, cool extremities, and/or impairment in mentation.⁶⁴

Patients with profile A (“dry-warm”) are compensated from a cardiac perspective, and presence of dyspnea in these patients should prompt evaluation for alternative causes of dyspnea. These patients, if not already taking a BB and ACEI/ARB should have these medications initiated and uptitrated as tolerated. Patients with profile B (“wet-warm”) generally require diuresis with or without vasodilation. Importantly, profile B patients generally do not require lowering or discontinuation of their background HF medications. Profile C (“wet-cold”) patients, in contrast, are generally a sicker group who may require inotropic support in order to achieve adequate diuresis and may require alteration in background HF medications such as holding or lowering BB and ACEI/ARB secondary to hypotension and/or WRF. This group may also require invasive hemodynamic monitoring, particularly if conventional therapy does not result in clinical improvement. Lastly, patients with profile L (“dry-cold”) are not common but include those with severe limitation in cardiovascular reserve.⁶⁴ Along with symptom relief, careful consideration to more definitive therapies for HF or advanced planning should be considered for these patients.

Diagnostic Testing

Several diagnostic tests are essential as part of a workup for AHF. All patient should have an electrocardiogram in the emergency department.⁵¹ Specific attention should be paid to signs of acute coronary syndromes and ischemia, QRS prolongation potentially indicating ischemia or ventricular dyssynchrony or dilatation, atrial and ventricular arrhythmias, and heart block. Chest radiography should be performed to assess for pulmonary congestion, cardiomegaly, pleural and pericardial effusions, and the other pulmonary processes that may be contributing factors, such as infection or pulmonary disease.⁵⁹ Initial laboratory testing should include complete blood count, serum electrolytes, renal indices, liver function testing, and urinalysis. Specific attention should be paid to the presence of hyponatremia, anemia, and renal function. Evidence of hepatic abnormalities may signify passive congestion and right ventricular dysfunction. Measurement of natriuretic peptides is recommended for patients in whom the diagnosis of AHF is undetermined or for additional prognostic information.⁵¹

Other noninvasive testing includes echocardiogram and, in appropriate cases, assessment for the presence of ischemia and myocardial viability. Echocardiography can be useful for determining severity of valvular abnormalities, assessment of diastolic dysfunction, quantification of pulmonary hypertension by estimation of pulmonary artery systolic pressures, and other hemodynamic parameters noninvasively.⁶⁵ Most importantly, echocardiogram is often useful in providing estimated right atrial pressure using the inferior vena cava size and collapse index. Transesophageal echocardiography may be performed in AHF patients who have inadequate quality of transthoracic images, in cases of severe valvular disease for additional quantification, in suspected endocarditis, in patients with suspected source of cardiac thrombus, and in congenital disease.⁶⁶ Although the role of handheld echocardiography is still evolving, this imaging modality may aid in estimation of LV filling pressures as well.⁶⁷ There are several modalities that can be used for noninvasive assessment of myocardial ischemia and/or viability in suitable patients. These include stress echocardiography, nuclear imaging (single-photon emission computed tomography and positron emission tomography), as well as cardiac magnetic resonance imaging.⁵¹

Pulmonary Artery (Swan-Ganz) Catheterization: Use and Indications

Pulmonary artery catheters (PACs) have historically been used as part of a “tailored therapy” approach in AHF. More recently, routine use of PAC-guided management has been discouraged.⁶⁸ The Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheterization and Effectiveness (ESCAPE) trial evaluated the use of PACs in patients hospitalized with HF in whom the use of a PAC was deemed to be of potential benefit but not essential.⁶⁹ The use of PACs was compared to clinical assessment alone in 433 patients with comparable baseline characteristics, including severe elevation in PCWP, diminished LV ejection fraction, and cardiac index 1.9 ± 0.6 L/min/m². The ESCAPE trial reported no differences between groups in the primary end point of days alive out of the hospital. Subgroup analysis showed a trend toward improved outcomes in the PAC group in the highest enrollment centers and a trend toward favor of PAC group for functional assessment (6-minute walk, Minnesota Living With Heart Failure testing), but also a trend toward higher adverse events.

It is important to note that the ESCAPE trial excluded patients whom physicians felt clearly required PAC for optimal management and involved physician investigators who were highly experienced in the evaluation and treatment of HF. With the bedside availability of astute physical exam skills, serum biomarkers and laboratory results, and point-of-care and advanced imaging techniques, PACs should be reserved for selective situations.⁷⁰ The ACC/AHA guidelines state that invasive hemodynamic monitoring should not be routinely used, but can play a role in carefully selected patients for whom hemodynamics are unclear or AHF symptoms are persistent.⁵¹

There may be a role for use of PACs, when performed by skilled operators knowledgeable in hemodynamic assessment, in identifying high-risk HF patients with persistent hemodynamic abnormalities, as well as in the trials for development of novel agents; however, improvement in education and guidelines are necessary to establish appropriate contemporary use of PACs.⁷¹

Biomarkers

Predicting individual risk in patients with AHF is challenging because of tremendous diversity in clinical presentation and underlying pathophysiology. The use of biomarkers can be additive in predicting risk after an AHF hospitalization⁷² and inform decision making regarding treatment.⁷³ Different biomarkers for use in AHF relate to the varied underlying pathophysiologic targets (Fig. 71–8).⁷³ Biomarkers reflecting myocardial stretch, the natriuretic peptides, are the most widely studied and used in AHF patients, and are powerful predictors of outcome.⁷⁴ Both atrial natriuretic peptide and B-type natriuretic peptide (BNP) have vasodilatory, natriuretic, and diuretic properties.⁷⁵ N-terminal pro–B-type natriuretic peptide (NT-proBNP) is a part of the precursor peptide BNP and has a longer half-life.⁷⁶ BNP and NT-proBNP are similar for diagnosis of AHF,⁷⁷ but NT-proBNP is superior in predicting clinical outcome.⁷⁸ Serial measurement may add to prognostic prediction. The Pro-BNP Outpatient Tailored Chronic Heart Failure Therapy (PROTECT) study showed that a 50% reduction in NT-proBNP correlated with a nearly 50% reduction in event rate.⁷⁹ Discharge value may be more predictive than admission value.^80,81

Several biomarkers may be reflective of the underlying neurohormonal imbalance of AHF, including activation of SNS, RAAS, and ADH release. Adrenomedullin (ADM), a peptide hormone secreted by myocardial tissue and endothelial cells, may be reflective of neurohormonal activation. ADM, because of a short half-life, is difficult to measure. Midregional pro-adrenomedullin (MRproADM) is more stable and thus more useful in the clinical setting.^82,83,84 Elevated levels of MRproADM are associated with adverse outcome in AHF patient.⁸⁵ High levels of vasopressin reflect severity of HF, but are difficult to measure as a result of instability and rapid clearance.⁸⁶ Copeptin is the C-terminal portion of the vasopressin prohormone and can be more easily measured. Copeptin is a marker for HF disease progression.⁸⁷ In patients with AHF, high levels of copeptin were predictive of mortality at 12 and 24 months.^88,89 Chromogranin A, a prohormone, is produced by adrenergic and neuroendocrine cells⁹⁰ and ultimately leads to vasoconstriction and catecholamine release.⁹¹ Chromogranin A is found in higher levels in patients with HF. Similar to copeptin, higher levels are associated with HF severity and can be predictive of mortality.⁹²

Another group of biomarkers is used to reflect the underlying pathophysiologic cardiac remodeling seen in HF, including myocardial hypertrophy, fibrosis, apoptosis, and necrosis. Soluble ST2, in response to volume overload and myocardial strain, is found in higher levels in patients with HF and is associated with poor prognosis.^93,94 ST2 levels can also add prognostic value to NT-proBNP and high-sensitivity troponin T.⁹⁵ Elevations in ST2 can aid in the identification of patients with decompensation and cardiac remodeling.⁹⁶ Galectin-3 is released by activated macrophages in response to cardiac stress and stimulates production of collagen, which leads to fibrosis and worsening ventricular dysfunction.^97,98,99 Galactin-3 has been shown to be an independent predictor of mortality¹⁰⁰ and rehospitalization¹⁰¹ in patients with AHF.

Downstream myocyte necrosis and apoptosis may be a result of increased myocardial wall stress, elevated LVEDP, and decreased myocardial perfusion.⁸⁴ Biomarkers of myocyte injury, such as troponins T and I, have been shown to add to prognosis in patients with AHF,¹⁰² although certain patient factors, such as renal failure, sex, age, and the presence of LV hypertrophy, can influence levels.¹⁰³ Patients with elevated troponin levels are generally sicker, with lower systolic blood pressure, lower ejection fractions, and higher inpatient mortality.¹⁰²

WRF and acute kidney injury are correlated with worsened mortality¹⁰⁴ and are prevalent in patients with AHF. There has been growing interest in biomarkers reflecting renal injury, such as cystatin C and neutrophil gelatinase-associated lipocalin, to predict prognosis. Cystatin C is a cysteine protease inhibitor released from all kidney cells and filtered by the glomerulus.^105,106 High levels of cystatin C are associated with AHF outcome.¹⁰⁷ Neutrophil gelatinase-associated lipocalin is secreted by kidney cells in response to acute kidney injury, preceding creatinine elevation,¹⁰⁸ and predicts 30-day mortality after AHF hospitalization.¹⁰⁸

AHF reflects a proinflammatory state. Levels of high-sensitivity CRP are elevated in AHF and associated with mortality.^109,110 Reliability of CRP levels may be diminished in patients with infections.¹¹¹

Biomarker utilization may add to risk prediction, although which combination and in what capacity remain largely undefined. In the Multinational Observational Cohort on Acute Heart Failure (MOCA) trial, a multinational cohort of AHF patients that studied the value of adding biomarkers to a clinical prediction model, MRproADM had the highest 30-day mortality, and the combination of elevated CRP and ST2 had the highest 1-year mortality in risk prediction. The Biomarkers in Acute Heart Failure (BACH) trial, a prospective multicenter study of patients presenting with dyspnea, showed MRproADM to be the best predictor of mortality compared with BNP and NT-proBNP, but showed that BNP was still a superior predictor of rehospitalization.¹¹² In AHF, BNP was superior to troponin I and high-sensitivity CRP for mortality prediction, but the combination improved statistical significance.¹⁰⁹

Identification and Management of Precipitating Factors

Patients requiring hospitalization for HF exacerbation can often be characterized as having a precipitating event (Table 71–5). Data from OPTIMIZE-HF represent the most comprehensive identification of precipitants leading to hospitalization for AHF,¹¹³ reporting that of the almost 50,000 patients included, roughly 60% had one or more identified precipitating factors. These precipitants independently predicted clinical outcomes, making early identification and targeted therapy critical. In the OPTIMIZE-HF registry, there were a number of factors found to precipitate hospital admission for AHF including ischemia or acute coronary syndromes, arrhythmia, uncontrolled hypertension, pneumonia, WRF, and noncompliance with diet or medications.¹¹³

Prompt identification and early medical management of precipitating factors in AHF are essential to improve outcomes of these patients.

Hospital management of AHF begins with immediate emergency department assessment followed by comprehensive in-hospital and predischarge management and planning (Table 71–6 and Fig. 71–9).^59,114 Two phases of AHF care have been described.^55,56 Phase I is the stabilization phase, often beginning with care in the emergency department, where the main goals include identification and improvement of signs and symptoms, hemodynamics, and correction of volume overload. Phase II continues into the hospitalization and through discharge planning. Primary goals include prevention of disease progression, focus on improving cardiac function, and following the evidence-based guidelines. Critical to phase I is early and appropriate identification of precipitating factors and determination of clinical profiles whereby therapies can be targeted appropriately based on patient selection. Paramount to phase II is appropriate incorporation of guideline-based therapy for both assessment and management, with careful attention to quality metrics to ensure safe discharge and optimize mortality benefit.

The presence of congestion is a leading factor in hospitalization for AHF. Multiple strategies for decongestion are used, including diuretics such as loop and thiazide and MRAs, vasodilators, ultrafiltration, vasopressin antagonists, and some novel agents.

Diuretics (Loop Diuretics, Thiazides, Mineralocorticoid Receptor Antagonists)

At the center of successful management of AHF is recognition and management of volume overload related to retention of salt and water in order to restore an euvolemic state. The mainstay of diuretic therapy is loop diuretics (furosemide, torsemide, bumetanide, and ethacrynic acid). In general, patients with AHF will require intravenous diuretics, with initial dose either equivalent to or exceeding their current home dose.⁵¹ Diuretic therapy should be delivered without delay because early intervention has been shown to improve outcome in observational studies.^78,115 The Diuretic Optimization Strategies Evaluation (DOSE) trial randomized AHF patients to low-dose (equal to daily diuretic dose) or high-dose (2.5 times total daily dose) furosemide, administered as either an intravenous bolus every 12 hours or as a continuous infusion. The study found no significant difference between strategies in the patient-reported global assessment of symptoms and no change in serum creatinine from baseline to 72 hours.¹¹⁶ However, there was a trend toward greater improvement in patients’ global assessment of symptoms in the high-dose versus the low-dose group. In addition, the high-dose strategy was associated with more favorable outcomes in some secondary measures, including lower levels of BNP, higher levels of diuresis, and relief of dyspnea.

Certain patients with refractory volume retention may require higher doses of their current loop diuretic, a switch to an alternative loop diuretic, or addition of a diuretic with a different mechanism of action such as a thiazide diuretic (ie, metolazone or chlorthalidone). Thiazides may be useful as an addition in patients on chronic loop diuretics because they may overcome loop diuretic resistance, but they should be used with caution because they can cause severe electrolyte disturbances such as profound hypokalemia.

MRAs such as spironolactone and eplerenone have established mortality benefit in chronic HFrEF patients in low doses, secondary to their antifibrotic properties and beneficial effects on cardiac remodeling.^117,118,119 Higher doses can augment diuresis by blocking salt-retaining aldosterone receptors.¹²⁰ Thus far, high-dose spironolactone has only been evaluated in one trial, where it led to decreased edema, orthopnea, and NT-proBNP levels.¹²¹ Further studies with MRA therapy in AHF are ongoing.

Mild elevations in blood urea nitrogen and creatinine should be monitored closely, but should not prompt lowering or withholding of diuretic therapy. Intravenous loop diuretics can, however, reduce glomerular filtration rate and create a cycle of worsening neurohormonal activation and electrolyte disturbances. Excessive or incorrect use can lead to hypotension or renal dysfunction. In some patients in whom volume overload is accompanied by pre-existing renal impairment, response to diuretic therapy may be attenuated. In most cases, adequate diuretic dosing improves renal function as a result of venous decongestion.¹²²

Patients requiring an augmented diuretic strategy require frequent reassessment and monitoring of volume status by physical exam, measurement of daily weight, measurement of supine and standing vital signs, and daily serum laboratory testing. Hospital discharge should only be considered once euvolemia has been achieved and an appropriate diuretic strategy for maintenance therapy has been established. Roughly 50% of patients are discharged from the hospital with residual congestion after which they are at considerably higher risk of early readmission and poor outcomes.¹²³

Vasopressin Antagonists

Approximately 25% of patients admitted with AHF have low serum sodium, which is a strong predictor for postdischarge mortality and readmission.¹²⁴ Solute-free water diuretics, or “aquaretics,” exert their effects through vasopressin antagonism. These agents include tolvaptan, conivaptan, and lixivaptan. They are most commonly used for the correction of hyponatremia, but have been evaluated in the setting of AHF. The Efficacy of Vasopressin Antagonism in Heart Failure Outcome Study With Tolvaptan (EVEREST) trial revealed a clinical benefit with vasopressin, with significant weight reduction and dyspnea scores at 24 hours and decreased volume retention at day 7. Despite this, there were no observed differences in hospitalization or mortality at 10-month follow-up.¹²⁵ Lixivaptan effectively removes fluid in NYHA class II and III chronic HF patients.¹²⁶ The ongoing Treatment of Hyponatremia Based on Lixivaptan in NYHA III/IV Cardiac Patient Evaluation Study will aim to define the role of this “vaptan” in AHF patients with hyponatremia.¹²⁷ Conivaptan differs from others because of intravenous delivery and has been shown in a small pilot study to improve diuresis when added to a loop diuretic regimen,¹²⁸ although larger scale studies are needed.

Ultrafiltration

If marked volume retention persists and more conservative strategies are limited by renal dysfunction, patients may require mechanical removal of volume with ultrafiltration (UF) or hemodialysis.^129,130 UF uses a volume removal strategy of moving water and small- to medium-weight solutes across a semipermeable membrane, using two peripheral venous catheters, thus avoiding complications of central venous access.¹³¹ UF machines are relatively simple to use and do not require the same level of care as hemodialysis.¹³² The Ultrafiltration Versus Intravenous Diuretics for Patients Hospitalized for Acute Decompensated Congestive Heart Failure trial was a 200-patient trial demonstrating the UF group to have greater weight loss and volume removal with no differences in dyspnea at 48 hours. By 90 days, the UF group had lower risk of HF hospitalization and fewer unscheduled clinic visits, without any difference in renal function or overall mortality.¹³³ The subsequent Cardiorenal Rescue Study in Acute Decompensated Heart Failure trial randomized 188 patients with AHF, WRF, and persistent intravascular congestion to receive either UF or high-dose loop diuretics (employing an aggressive strategy of addition of thiazide diuretics, inotropic agents, or vasodilators, as needed), and found no differences in mean weight loss at 96 hours. The UF group had more serious adverse events, mainly bleeding and worsened serum creatinine levels. As a result of uncertainty surrounding the results of these trials, a subsequent 800-patient trial, the Study of Heart Failure Hospitalizations After Aquapheresis Therapy Compared to Intravenous Diuretic Therapy, was undertaken but terminated prematurely secondary to slow patient recruitment.¹³⁴ The ACC/AHA guidelines currently state that UF is reasonable in certain patients with refractory congestion once other strategies have failed.⁵¹

Vasodilator Therapy

In patients with severe volume overload in the absence of systemic hypotension, intravenous vasodilators such as nitroglycerin, nitroprusside, or nesiritide may play a beneficial role when added to diuretic therapy.⁵¹ These agents may be particularly suited in patients with adequate blood pressure experiencing refractory congestion with inadequate diuretic response. Intravenous nitroglycerin, primarily a venodilator, lowers preload, thereby relieving pulmonary congestion. Patients with hypertension, coronary ischemia, or severe mitral regurgitation are considered good candidates for nitroglycerin therapy.¹³⁵ Limitations include rapid development of tachyphylaxis, leading to drug resistance even at high doses.¹³⁶ Sodium nitroprusside is a balanced venous and arterial vasodilator that also provides some pulmonary vasodilation. Sodium nitroprusside can be used in markedly congested patients with hypertension or severe mitral regurgitation, although use is often limited by potential for systemic hypotension and potential thiocyanate toxicity, especially in patients with renal insufficiency.¹³⁵ Nesiritide reduces LV filling pressures and relieves dyspnea, but may lead to hypotension. Nesiritide was shown to improve early hemodynamics in the Vasodilation in the Management of Acute Heart Failure trial.¹³⁷ However, the subsequent Acute Study of Clinical Effectiveness of Nesiritide in Decompensated Heart Failure trial failed to reach statistical significance for either relief of dyspnea or HF readmission or death at 30 days compared with placebo.¹³⁸ Furthermore, the Renal Optimization Strategies Evaluation in Acute Heart Failure trial, with low-dose nesiritide versus low-dose dopamine in patients with AHF and renal failure, also failed to show improvements in urine output, renal function, rehospitalization, HF symptoms, or mortality, and showed nesiritide to be associated with more symptomatic hypotension than placebo.¹³⁹

The endothelin receptor antagonist tezosentan was assessed in the Value of Endothelin Receptor Inhibition With Tezosentan in Acute Heart Failure Studies (VERITAS) study, which demonstrated no difference in tezosentan versus placebo in dyspnea, worsening HF, or death.¹⁴⁰ The Relaxin in Acute Heart Failure trial compared 48-hour serelaxin infusion versus placebo in patients presenting within 16 hours for AHF.¹⁴¹ Serelaxin improved patient-reported dyspnea on a visual analog scale, but not on a Likert scale, and was associated with improvement in 6-month cardiovascular mortality risk.¹⁴² Similarly, 24-hour infusion of ularitide was evaluated in the Safety and Efficacy of an Intravenous Placebo-Controlled Randomized Infusion of Ularitide in Symptomatic, Decompensated Chronic Heart Failure I and II trials.^143,144 Both serelaxin and ularitide are currently being investigated in ongoing phase III studies.

According to ACC/AHA guidelines, the use of vasodilators cannot be generalized because of lack of definitive data, but may play a role in specific situations, including relief, augmentation of blood pressure control complicating HF, and as an aid for the improvement in hemodynamic abnormalities while transitioning to oral medications.⁵¹

Sodium and Fluid Restriction

Historically, based on neurohormonal imbalances leading to impaired excretion of sodium and water, HF patients have been advised to limit intake of both. Recent studies have questioned that high sodium intake correlated with development of AHF.¹⁴⁵ In the inpatient setting, sodium restriction may or may not be beneficial. In AHF patients randomized to a fluid- (800 mL/d) or sodium-restricted (800 mg/d) diet for 3 days of hospitalization, there were no differences in length of stay, renal function, body weight, congestion, or readmission, but perceived thirst was higher in the treatment group.¹⁴⁶ In a study of combination high-dose furosemide with high-dose hypertonic saline, patients randomized to the combination strategy versus high-dose diuretic alone had improved diuresis, natriuresis, and renal function, with lower readmission rates and mortality.¹⁴⁷ The current ACC/AHA guidelines suggest that sodium restriction is reasonable in patients with symptomatic HF, and fluid restriction of 1.5 to 2 L/d can be considered in stage D patients, particularly with concurrent hyponatremia.⁵¹

Assessment of Perfusion and Hypotension

One of the strongest predictors of inpatient and postdischarge mortality in HF patients is low blood pressure.¹⁴⁸ Approximately 25% of patients at hospital presentation have low blood pressure, defined as systolic blood pressure ≤ 120 mm Hg, among which a subset has clinical signs and symptoms of low CO and systemic hypoperfusion. AHF patients with blood pressure < 110 mm Hg have an in-hospital mortality of 12% versus 4% for those who are normotensive.¹⁴⁹ OPTIMIZE-HF found in-hospital all-cause mortality to be highest for those with the lowest quartile blood pressure (defined as systolic blood pressure < 120 mm Hg).¹⁴⁸ EVEREST found patients with low blood pressure to be at high risk for rehospitalization from immediately after discharge to at least 18 months after discharge.¹⁵⁰ All oral agents known to improve long-term mortality in HF patients reduce blood pressure, greatly limiting use in this subset of patients for whom the therapeutic target is increased CO.¹⁵¹

Assessment of systemic perfusion is an essential component in management of AHF. A small subset of patients presents with a clinical scenario of low CO, manifested by low blood pressure (with or without congestion) accompanied by signs of diminished end-organ perfusion. Clinical manifestations may include cool or clammy skin, cool extremities, decreased urine output, and alterations in mental status. In patients considered to be “low output,” consideration should be given to use of inotropic agents, such as dopamine, dobutamine, and milrinone (Table 71–7). Milrinone, a phosphodiesterase-3 inhibitor, decreases the degradation of cyclic adenosine monophosphate, thus increasing protein kinase A, leading to phosphorylation of multiple myocardial targets downstream augmenting contractility. Milrinone is termed an inodilator because it effects contractility and decreases systemic and pulmonary vascular resistance. Because of its ability to improve pulmonary vascular resistance, milrinone administration may produce reversibility of marked pulmonary hypertension associated with poor CO.¹⁵² The mechanism of action of milrinone is independent of β-adrenergic receptors, which are downregulated in HF. Because of this, the effect of the drug is independent of concomitant use of β-blocking agents and may influence a clinician’s choice to choose milrinone over dobutamine. Milrinone has a half-life of roughly 2.4 hours and is renally cleared. Outcomes of a Prospective Trial of Intravenous Milrinone for Exacerbations of Chronic Heart Failure¹⁵³ evaluated the use of milrinone (0.5 μg/kg/min) versus placebo within 48 hours of hospital admission in all AHF patients not limited to the low-output group. Milrinone use did not result in decreased duration of hospitalization and was associated with a nonsignificant higher mortality both in hospital and after discharge, as well as higher rates of new-onset atrial arrhythmias and sustained hypotension requiring intervention. Thus, routine use of milrinone is not indicated in AHF.

Dobutamine is a sympathomimetic agent whose mechanism of action is through stimulation of β₁ receptors. The effect of dobutamine may be attenuated in patients who are concurrently taking BBs. Because it does not have direct pulmonary vasodilating properties, dobutamine typically does not affect pulmonary vascular resistance. Dobutamine has a half-life of 2 minutes with largely hepatic clearance. Both milrinone and dobutamine can cause a tachycardic response and are often arrhythmogenic, portending a higher risk of death. Observational analysis from the ADHERE registry suggested increased mortality with inotrope use in patients hospitalized with AHF, mostly in those with normal or elevated blood pressure.¹⁵⁴ ACC/AHA guidelines suggest that short-term inotropic support may be considered only for those patient with hypotension and signs of low CO, and long-term inotrope use may be considered for palliative treatment in stage D patient who are not candidates for advanced therapies. Guidelines also warn of harm in routine use of such parenteral agents.⁵¹

Levosimendan is a calcium sensitizer whose inotropic effect relates to its binding to calcium-saturated troponin C in the myocardial thin filament, which results in prolongation of actin-myosin coupling. In contrast to other inotropic agents, levosimendan does not cause an increase in myocardial oxygen demand. Additionally, levosimendan also has vasodilatory, anti-inflammatory, and antiapoptotic properties.¹⁵⁵ In 2013, an expert panel reviewed potential uses for levosimendan in advanced HF patients and concluded there were several clinical scenarios in which the drug may be suitable to maintain clinical stability, although mortality data are lacking.¹⁵⁶ Levosimendan may also have a role in various types of cardiogenic shock, including fulminant myocarditis, post–myocardial infarction, and Takotsubo cardiomyopathy, although prospective study is needed.¹⁵⁵ The second Randomized Multicenter Evaluation of Intravenous Levosimendan Efficacy trial¹⁵⁷ and Survival of Patients With Acute Heart Failure in Need of Intravenous Inotrope Support trial¹⁵⁸ did not show an overall mortality benefit in levosimendan compared with placebo or dobutamine, respectively, and currently the drug is not approved for use in AHF in the United States. The drug may have a role in select patients undergoing cardiac surgery, which is currently being studied in the Levosimendan in Patients With Left Ventricular Systolic Dysfunction Undergoing Cardiac Surgery on Cardiopulmonary Bypass trial (NCT02025621).

Neurohormonal Modulating Therapies

Numerous clinical trials have suggested that patients with chronic HFrEF respond as favorably to oral neurohormonal modulating therapies.⁵¹ There is evidence to suggest that continuation or limited change in oral therapy in AHF is possible and positively affects outcomes.¹⁵⁹ In patients hospitalized with AHF with reduced ejection fraction not already being treated with oral therapies known to improve outcomes, such as ACEI/ARBs, BBs, and MRAs, these agents should be initiated prior to discharge.⁵¹ The decision to initiate (or reinitiate) BBs should be based on volume status optimization, generally after intravenous diuretics, vasodilators, and inotropic medications have been discontinued and hemodynamic stability has been established. Patients hospitalized with hypertension should have their medications uptitrated during hospitalization whenever possible. Importantly, the majority of patients with AHF should not have their oral HF therapy held because studies suggest that continuation is associated with improved outcomes and is well tolerated.^160,161

There are certain scenarios during which therapy may require discontinuation or adjustment.¹⁶² Patients with marked volume overload with marginal hemodynamics may require temporary disruption or lowering of BB dosing until further euvolemia is achieved. Patients with WRF, azotemia, or hyperkalemia may require cessation or downtitration of ACEI and/or aldosterone antagonist therapies, pending improvement in renal function. When starting or restarting ACEIs and BBs, initial dosing should be low, with careful attention to intolerance. Hospitalization for AHF often presents a unique opportunity to optimize oral HF medications in a more controlled setting with careful attention to blood pressure and renal function, which is often difficult to do with tenuous outpatients.

There are certain hemodynamic conditions that require additional hemodynamic support. These clinical situations vary widely but result in compromised myocardial oxygen supply or demand. With acute myocardial infarction, reduced LV contractile function, increasing myocardial oxygen demand, and acute elevations in LV end-diastolic volume and pressure are coupled with diminished coronary blood flow. Cardiogenic shock is defined as systemic tissue hypoperfusion caused by inadequate CO despite adequate circulatory volume and filling pressure. Hemodynamic criteria include systolic blood pressure < 90 mm Hg for > 30 minutes, a drop in mean arterial pressure > 30 mm Hg below baseline with cardiac index < 1.8 L/min/m² without hemodynamic support, and a PCWP > 15 mm Hg.^163,164 For patients with advanced HF, the Interagency Registry for Mechanically Assisted Circulatory Support has defined seven clinical profiles with respect to acuity of hemodynamic deterioration, where profiles 1 and 2 refer to patients who either have acutely decompensated or are rapidly failing ionotropic therapy.¹⁶⁵ A consensus statement was published in 2015 to address the use of mechanical circulatory support in cardiovascular care.¹⁶⁶ The optimal timing for placement of mechanical circulatory support remains a source of debate, but the landscape of available options has evolved. Many factors influence the type of mechanical circulatory support to use, including the degree of hemodynamic instability, technical considerations, and acuity of need for implantation.

Intra-aortic balloon pump remains the most used temporary circulatory support device as a result of its widespread availability and familiarity. The intra-aortic balloon pump increases diastolic blood pressure, decreases myocardial oxygen consumption, improves coronary arterial perfusion, and provides up to 500 mL of additional CO. For patients requiring additional hemodynamic support, left atrium to aorta and left ventricle to aorta assist devices can be considered. The Tandem heart is currently the only left atrium to aorta assist device commercially available and is inserted percutaneously via a trans-septal approach to pump extracorporeal blood from the left atrium to the iliofemoral artery, bypassing the left ventricle. The TandemHeart (TandemLife, Pittsburgh, PA) reduces LV preload, filling pressures, wall stress, and myocardial oxygen demand and, depending on cannula size, can provide up to 5 L of circulatory blood flow. Adequate right ventricular function or the use of simultaneous right ventricular assist device is necessary to maintain adequate LA preload. The Impella (Abiomed, Danvers, MA) is a nonpulsatile axial flow device that propels blood from the left ventricle, to the aorta and, depending on the specific device, can provide from 2.5 L/min (Impella 2.5) up to 5 L/min (Impella 5.0) of circulating blood volume. Similar to the TandemHeart, Impella decreases myocardial oxygen consumption, improves mean arterial blood pressure, and reduces PCWP. This device also relies on adequate preload for optimal functioning. Extracorporeal membrane oxygenation provides cardiopulmonary support either as veno-veno or veno-arterial support, with veno-arterial extracorporeal membrane oxygenation used as the mechanical circulatory support system of choice in patients requiring full biventricular support secondary to cardiogenic shock and inadequate oxygenation. Although veno-arterial extracorporeal membrane oxygenation can provide flows exceeding 6 L/min, the system often requires additional ventricular unloading (either by intra-aortic balloon pump or Impella) to reduced ventricular wall stress and myocardial oxygen demand. Bleeding, infection, limb ischemia, and hemolysis can complicate any of the aforementioned mechanical circulatory support devices. Extracorporeal devices can be used for short-term circulatory support in patients who are expected to recover from a major cardiac insult (eg, myocardial ischemia, postcardiotomy shock, or fulminant myocarditis).¹⁶⁶

Patients hospitalized with HF carry an increased risk of thromboembolic events and should receive prophylactic anticoagulation unless contraindicated.¹⁶⁷ Whenever possible, attempts should be made to transition patients from intravenous to oral medications, with careful attention to hemodynamic compromise once this occurs. Treating clinicians should assess and discuss overall prognosis with patients and their caregivers, once functional status and precipitating factors have been addressed.⁵¹ When indicated, discussions regarding advanced therapies and/or end-of-life care should ensue¹³³ (Table 71–8).

Many registries suggest that patients with AHF are commonly discharged before euvolemia is achieved, without optimal blood pressure control and without optimal use of life-saving medications such as ACEIs/ARBs and BBs.⁵⁹ To optimize posthospitalization outcomes, clinicians should follow published guidelines.⁵¹ A critical performance measure set forth for HF hospitalization involves care coordination and transition of care, which includes written discharge instructions or educational materials provided to patients explaining discharge medications, activity level, diet, weight monitoring, follow-up visit instructions, and what to do if symptoms worsen.¹⁶⁸ Patients admitted with AHF should be evaluated prior to discharge for appropriateness for devices, including cardiac resynchronization therapy and implantable cardioverter-defibrillator.¹⁶⁹ Consideration should also be given to candidacy for implantable pressure sensors such as CardioMEMS (St. Jude Medical, St. Paul, MN), which may aid in decreasing rehospitalization compared with clinical assessment alone.¹⁷⁰ Patients should be provided with referral to a tertiary care centers with expertise in mechanical device support and/or orthotopic heart transplantation and other advanced surgical options as indicated for advanced disease.⁵¹

Early postdischarge assessment should include careful attention to congestion, the leading cause of readmission, optimizing diuretic doses, and other guideline-recommended drug- and device-based therapies, including uptitration of medications to recommended doses.⁵⁹ Recent data suggest that follow-up with a physician within 7 days of discharge¹⁷¹ and follow-up with a cardiologist¹⁷² can lower 30-day mortality. HF hospitalization is a well-recognized independent risk factor for mortality in patient with chronic HF, and the postdischarge period a particularly vulnerable phase for patients.¹¹⁴ Comprehensive discharge planning ensuring compliance with guideline-recommended therapy can reduce readmissions and improve patient outcomes.¹⁷³

HF is the number one cause of hospitalization among Medicare beneficiaries, and each hospital readmission portends worse risk of death than previous admission. Nearly 25% of patients are rehospitalized within 30 days, and 50% of HF patients are rehospitalized within 6 months of index hospitalization.^18,19 In observational registry data, mortality was reported to be approximately 40%, and readmission rate was nearly 70% at 1 year after index AHF admission.¹⁷⁴ With the aging population, HF rates and rehospitalizations are estimated to rise significantly. Several risk factors for rehospitalization have been identified, including but not limited to advanced age; ischemic etiology; low systolic blood pressure; increased heart rate; higher NYHA class at discharge; laboratory abnormalities such as low admission hemoglobin, admission and discharge creatinine, and increased blood urea nitrogen; other testing abnormalities such as prolonged QRS on electrocardiogram; echocardiographic parameters such as lower ejection fraction; higher systolic pulmonary artery pressure; and right ventricular tissue Doppler imaging.¹⁸ Similarly, natriuretic peptide levels at admission and on discharge predict readmission risk, and postdischarge measurement also guides further risk stratification.¹⁸ Both cardiac and noncardiac comorbidities increase risk of rehospitalization, as do patient-reported quality of life scores and psychosocial factors.¹⁸ In addition, prolonged index hospitalization length of stay predicts higher risk of 30-day all-cause hospitalization.¹⁷⁵

Currently, the 30-day postdischarge hospitalization rate is a metric that is associated with financial penalties by Medicare. Identification of low- and high-risk patient is necessary to risk stratify patients for likelihood of readmission in order to optimize prevention interventions. However, although several risk prediction models have been described for readmission risk, knowledge gaps in pathophysiology and heterogeneity of the patient population render these scores less accurate than risk prediction models for mortality. A recent scientific statement by the AHA summarizes available transitional care models with respect to interventions and outcomes, highlighting the need for further study in order to best integrate this approach. Key components included patient education prior to discharge focused on “teach back” to ensure patient engagement and understanding, early postdischarge follow-up and reassessment, and effective communication between in-hospital and outpatient providers.¹⁷⁶ Several strategies appear to be effective, especially when used in combination, to reduce HF readmissions, including partnering of inpatient teams with local physicians and local hospitals, medication reconciliation, assuring follow-up appointments before discharge, having discharge summaries sent to primary care physicians, and assigning team members to follow up on pending test results after discharge.¹⁷⁷ A “three-phase terrain” of HF readmission describes the predisposition of patients to early rehospitalization following admission and highlights this vulnerable period in the trajectory of a hospitalized HF patient, underscoring the need for tailored strategies targeted at preventing readmission.¹⁷⁸