several days ago with severe, progressive, and debilitating symptoms of HF. His family history is significant in that

his father and two brothers died of heart attacks shortly

after the age of 40. L.M. has a 5-year history of HF that

is symptomatic despite treatment with furosemide 40 mg

every day, lisinopril 10 mg every day, carvedilol 12.5 mg

BID, digoxin 0.125 mg every day, and spironolactone

25 mg every day. His last echocardiogram showed an EF

of 25%. He has a past medical history of HTN, CAD, and

HF. During the last week, L.M.’s DOE became progressively worse, and he was mostly confined to bed because of

477Heart Failure Chapter 19

extreme fatigue. He wakes once or twice nightly with PND.

Physical examination on this admission revealed significant

jugular venous distension, bilateral rales, hepatomegaly,

and 3+ peripheral edema. Chest radiograph revealed

cardiomegaly and pulmonary congestion. His BP was

154/100 mm Hg and his pulse was 105 beats/minute.

Laboratory test results include the following:

Serum Na, 134 mg/dL

K, 4.3 mg/dL

BUN, 15 mg/dL

SCr, 1.3 mg/dL

Glucose, 90 mg/dL

BNP, 944 pg/mL

Hemoglobin, hematocrit, troponin, and liver enzymes were

all within normal limits. L.M. was admitted to the hospital with ADHF. What factors are used to stratify risk for

in-hospital mortality and morbidity?

ADHF is associated with high morbidity and mortality. It is

the leading reason for hospitalizations among the elderly. Postdischarge mortality among patients with HF is 11% at 30 days

and 37% at 1 year.10 The most common precipitants of HF hospitalization are noncompliance (dietary or medication), acute

myocardial ischemia, uncontrolled comorbidities (such as HTN,

diabetes mellitus, chronic kidney disease, arrhythmias), and inappropriate prescribing (e.g., recent addition of negative inotropic

agents or NSAIDs). Regardless of the precipitating event, the

common pathophysiologic state that perpetuates the progression of HF is extremely complex. A cascade of hemodynamic and

neurohormonal derangements provoke activation of adrenergic

systems and RAAS, which get overwhelmed, leading to volume

overload and hypoperfusion—classic symptoms of ADHF (see

a detailed discussion in the Pathophysiology section). According to the ADHF National Registry, which is one of the largest

multicenter, observational registries, nearly half of the patients

who present with ADHF have preserved EF (>50%).297 Because

so few targeted therapies are available for HFPEF, ADHF in this

subset of patients poses a unique challenge.

In-hospital mortality remains as high as 20% for patients who

present with elevated BUN and creatinine, and low systolic BP

(SCr >2.75 mg/dL, systolic BP <115 mm Hg, and BUN >43

mg/dL).287 Patients presenting with hyponatremia, defined as a

serum sodium level less than 135 mEq/L, have the worst outcomes and are more likely to receive inotropic agents. After discharge, patients who continue to be hyponatremic have an 8%

risk of rehospitalization per 3 mEq/L decrease in serum sodium.

Uric acid greater than 7 mg/dL in men and greater than 6 mg/dL in

women is associated with a high admission rate for HF. Admission

and discharge levels of BNP are also helpful in predicting rehospitalizations. According to some studies a greater than 30% to

40% reduction in BNP levels during hospitalization may result in

improved outcomes. Other biomarkers such as cardiac troponin

levels, C-reactive protein, and apolipoprotein A-I levels are all

linked to HF readmissions.11 Thus, all these factors can be used

to stratify risk for in-hospital mortality.

Another approach to risk stratification and therapeutic decision making is based on hemodynamic profiles. Most patients

with acute HF can be classified into one of four hemodynamic profiles using relatively simple assessment techniques

(Fig. 19-8).298–301 When using this scheme, patients are assessed

Normal

Hypovolemic shock Cardiogenic shock

Pulmonary edema

Indicators of low organ perfusion:

Hypotension: SBP <100 mm Hg

↑ Peripheral vascular resistance

↓ Cardiac output

Weak peripheral pulse

Cool extremities

↓ Mental alertness

↑ BUN or SCr

↓ Urine flow

Indicators of high filling pressure:

↑ Right atrial pressure

↑ Pulmonary pressure

Orthopnea

Jugular venous distension

Third heart sound

Peripheral edema

Ascites

Reduced blood pressure: Inotropic

agents or vasopressors

Fluid administration

Inotropic agents

I II

III IV

Cl

(L/minute/m2)

Pulmonary Congestion

Normal Perfusion

FIGURE 19-8 Hemodynamic profile of

acute heart failure. BUN, blood area

nitrogen; CI, cardiac index; NTG,

nitroglycerin; PCWP, pulmonary capillary

wedge pressure; SBP, systolic blood

pressure; SCr, serum creatinine. Adapted

with permission from Forrester JS et al.

Correlative classification of clinical and

hemodynamic function after acute

myocardial infarction. Am J Cardiol.

1977;39:137.

478 Section 2 Cardiac and Vascular Disorders

for the presence or absence of elevated venous filling pressures

(“wet” vs. “dry” patients) and adequacy of vital organ perfusion

(“warm” vs. “cold” patients). Elevated filling pressure can be

assessed at the patient’s bedside by observing orthopnea, jugular venous distension, the presence of a third heart sound (S3),

peripheral edema, and ascites. Presence or absence of rales on

auscultation is not considered a reliable indicator.301 Hypotension, weak peripheral pulse, a narrow pulse pressure, cool forearms and legs, decreased mental alertness, and rising BUN and

SCr are indicators of decreased organ perfusion. In one series,

67% of patients admitted to the hospital with a low LVEF and

class IV HF symptoms were classified as “wet and warm,” with

28% assessed as “cold and wet,” and only 5% as “cold and dry.”298

Few, if any, patients were in the “warm and dry” category because

this is the status the clinician is trying to achieve. Continuous BP,

ECG, urine flow, and pulse oximetry measurements are standard

noninvasive monitoring for all patients. Invasive hemodynamic

monitoring is used in critically ill patients when more precise

measurements of filling pressure (e.g., right atrial or pulmonary

artery pressure), SVR, and CO or cardiac index are desired. The

goals are to achieve a right atrial pressure of less than 5 to

8 mm Hg, pulmonary artery pressure of less than 25/10 mm

Hg, pulmonary artery wedge pressure of 12 to 16 mm Hg or

less, an SVR of 900 to 1,400 dyne · s/m5 and a cardiac index of

2.8 to 4.2 L/minute/m2 (see Chapter 22, Shock, for more detailed

discussion of hemodynamic monitoring).

CASE 19-4, QUESTION 2: From his clinical presentation,

what is L.M.’s hemodynamic profile? What are therapeutic

goals for L.M.?

L.M.’s DOE, rales, peripheral edema, PND, and jugular

venous pressure are consistent with ADHF and volume overload.

L.M. does not have signs of hypoperfusion (does not have cool

extremities and acute elevations in SCr). Therefore, his hemodynamic profile based on physical examination is “warm and

wet” (subset II, Fig. 19-8). According to the guidelines, a rapid

diagnosis of ADHF is necessary to initiate appropriate treatment

and should be based on signs and symptoms. In patients who

have established HF, precipitating factors should be identified

and addressed. Also, at the time of discharge, medications that

have been shown to decrease morbidity and mortality in HF

should be optimized. BNP levels should be considered when the

diagnosis is uncertain and, in conjunction with signs and symptoms of HF, can be helpful in differentiating between cardiac and

noncardiac causes.

Because volume overload is central to the pathophysiology of

most episodes of ADHF, the primary goal is to relieve congestion

in patients in subset II (i.e., warm and wet). However, diuretics

should be cautiously used in subset IV to avoid decreases in PCW

less than 15 mm Hg as this can compromise preload and reduce

CO. Despite diuretics being the mainstay therapy in reducing

volume overload, their routine use in ADHF is associated with

worsening mortality. Mortality rates are higher in patients who

are receiving chronic diuretic therapy compared with those not

receiving diuretics (3.3% vs. 2.7%, respectively). Although these

findings are based on the analysis of retrospective data302 versus prospective randomized trials, nevertheless, they suggest an

association between mortality and diuretic use in patients with

ADHF. (See Case 19-1, Question 7, for diuretic dosing.)

ULTRAFILTRATION

Ultrafiltration is an alternative approach for treating volume overload and congestion. It involves the use of a small, portable device

to rapidly remove fluid (up to 500 mL/hour). Typically, ultrafiltration for HF is reserved for patients with renal failure or those

unresponsive to diuretics. Small studies303,304 in ADHF patients

have shown that peripheral venovenous ultrafiltration results

in greater weight loss and fluid removal, and reduced hospital

length of stay and hospital readmissions compared with medical therapy. One study also reported significant reductions in

neurohormonal activation and no significant changes in SCr

or electrolytes with ultrafiltration.305 The Ultrafiltration versus Intravenous Diuretics for Patients Hospitalized for Acute

Decompensated Heart Failure (UNLOAD trial)306 randomly

assigned 200 patients to either ultrafiltration or aggressive IV

diuretic therapy. The study showed that ultrafiltration compared

with diuretics alone significantly improved weight loss at 48 hours

(5.0 vs. 3.1 kg; p <0.001), which decreased the need for vasoactive drugs (3% vs. 13%; p = 0.02) and reduced 90-day hospital

readmission (18% vs. 32%; p = 0.02). Ultrafiltration is now a

class IIa recommendation for patients with refractory HF not

responsive to medical therapy. Patients with fluid overload and

some degree of renal insufficiency and those refractory to diuretic

therapy might make good candidates for ultrafiltration.1,2

Intravenous Vasodilators

CASE 19-4, QUESTION 3: L.M. received a 40-mg IV

furosemide dose with no improvement in symptoms. Then

he received an 80-mg IV dose with only marginal resolution

of symptoms. The decision is made to start IV vasodilators.

What is the role of vasodilator therapy in someone with

ADHF such as L.M.?

The guidelines1 recommend the addition of vasodilators in

conjunction with diuretics to reduce congestion in patients with

fluid overload. In the presence of asymptomatic hypotension, IV

NTG, nitroprusside, or nesiritide may be considered cautiously in

combination with diuretics for rapid improvement of ADHF. In

the presence of low SVR, arterial vasodilators (e.g., nitroprusside,

high-dose NTG, nesiritide) can further compromise perfusion,

especially in patients who have pre-existing hypotension.

NITROPRUSSIDE

Nitroprusside dilates both arterial and venous vessels; therefore,

it has the theoretical advantage of decreasing both afterload and

preload. Nitroprusside is potentially of value in severely congested patients with HTN or severe mitral valve regurgitation

complicating LV dysfunction. Its major disadvantages include

risk of hypotension that can cause a decrease in CO and reflex

tachycardia; coronary steal (in ischemic patients); and accumulation of toxic metabolites, including thiocyanate toxicity (see

Chapter 21, Hypertensive Crises). Cyanide toxicity is most likely

seen in renal insufficiency and in patients who receive more than

4 mcg/kg/minute of nitroprusside for more than 48 hours. When

stopping nitroprusside therapy, a slow taper is recommended

because a rebound increase in HF has been observed 10 to

30 minutes after drug withdrawal. (For a more thorough discussion of nitroprusside use, see Chapter 21, Hypertensive Crises.)

In addition, nitroprusside must be given by continuous IV

infusion, which necessitates arterial line placement and intensive

care unit admission in most situations, and it is unstable if exposed

to heat and light after reconstitution.

INTRAVENOUS NITROGLYCERIN

As the prototype of all nitrates, NTG primarily dilates the venous

capacitance vessels with only a slight effect on the arterial bed.

Patients with acute MI and pulmonary edema are often considered as ideal candidates for the use of IV NTG. It is hoped

that the resulting reduction in LV filling pressure (preload) will

479Heart Failure Chapter 19

reduce PCWP to less than 18 mm Hg. Because nitrates have minimal or no effect on afterload, CO will likely remain unchanged

or increase only slightly. NTG actually can decrease CO in some

patients by reducing the LV filling pressure to less than 15 mm Hg.

NTG is generally initiated at 10 mcg/minute and increased by

increments of 10 to 20 mcg, until the patient’s symptoms are

improved or PCWP is less than 16 mm Hg. The most common

side effect is headache, which can be treated with analgesics and

often resolves after continuous therapy. Tachyphylaxis to NTG

occurs as early as 1 to 2 hours after initiation of therapy. Approximately 20% of HF patients exhibit resistance to high doses of

NTG.1 Nitroprusside is a better choice for patients such as L.M.

because he has elevated BP.

NESIRITIDE

Nesiritide is recombinantly produced human B-type natriuretic

peptide containing the same 32 amino acids as native human

BNP.40,41 (The pharmacologic activity of the natriuretic peptides

was reviewed in the Pathogenesis section.) BNP binds to guanylate cyclase receptors on vascular smooth muscle (the BNP

receptor), leading to expression of cyclic guanosine monophosphate and subsequent vasodilation. Other actions include inhibition of ACE, sympathetic outflow, and ET-1. Peripheral and

coronary dilation coupled with improved RBF and increased

glomerular filtration all contribute to the beneficial effects of

nesiritide. Metabolic clearance of nesiritide is by a combination

of binding to cell surfaces with subsequent cellular internalization and lysosomal proteolysis as well as proteolytic cleavage by

endopeptidase (e.g., neutral endopeptidase). It undergoes only

minimal renal clearance. The mean elimination half-life is 8 to 22

minutes (mean, 18 minutes), necessitating a constant IV infusion.

In clinical trials of hospitalized patients with severe HF, nesiritide produced hemodynamic effects and reduction in dyspnea

scores comparable to NTG when used in combination with

IV diuretics and either dopamine or dobutamine.40,41 Dosedependent hypotension is the most common side effect with

nesiritide, reported in 11% to 32% of patients. In some trials, nesiritide caused a higher incidence or longer duration of hypotension

than NTG, whereas in other trials the incidence of hypotension was similar. The incidence of PVC and nonsustained ventricular tachycardia is less with nesiritide than with dopamine,

dobutamine, or milrinone. Other side effects include headache,

abdominal pain, nausea, anxiety, bradycardia, and leg cramps.

Because of high cost, the use of nesiritide is generally restricted

to those patients with acute HF exacerbations who are fluid

overloaded and have a PCWP greater than 18 to 20 mm Hg

despite high doses of diuretics and IV NTG. In contrast to NTG,

nesiritide has natriuretic properties that are additive to those

of the loop diuretics. It should be avoided in patients with a

systolic BP less than 90 to 100 mg Hg or in cases of cardiogenic

shock. Dobutamine or milrinone should be added or substituted

in hypotensive patients or those with a cardiac index of less than

2.2 L/minute/m2.

The IV infusion of nesiritide is prepared by diluting the contents of a 1.5-mg vial to 6 mcg/mL in 250 mL of 5% dextrose

or 0.9% NaCl. An initial loading dose of 2 mcg/kg is sometimes

given intravenously for 60 seconds, followed by a continuous IV

infusion at a rate of 0.01 mcg/kg/minute. The desired response is

a reduction of PCWP of 5 to 10 mm Hg at 15 minutes. The dose

can be increased in 0.005-mcg/kg/minute increments at 3-hour

intervals to a maximum of 0.03 mcg/kg/minute. Dosage should

be titrated to a PCWP less than 18 mm Hg and a systolic BP

greater than 90 mm Hg.

Although nesiritide is indicated in the treatment of ADHF,

concerns have been raised about its safety. A meta-analysis of

randomized, controlled trials of nesiritide in ADHF suggests

that nesiritide use may be associated with worsening renal function and increased mortality.307,308 The conclusions of the metaanalysis have been criticized, however. For example, the Vasodilation in the Management of Acute Congestive Heart Failure

(VMAC) study that was included in the meta-analysis was not

designed to evaluate renal end points, and therefore the inclusion

of the renal effects from this study may not be appropriate.302,309

Also, differences in baseline characteristics of the treatment

groups may have contributed to an increased risk of 30-day mortality seen in those patients treated with nesiritide.

The efficacy and safety of nesiritide have also been evaluated

in the outpatient setting. The Follow-up Serial Infusions of Nesiritide (FUSION I) trial310 was a pilot trial that included 202 patients

with NYHA class III and IV who had been hospitalized for ADHF

at least twice within the preceding year and once in the preceding month. Patients were randomly assigned to receive usual care

with or without open-label nesiritide. A subgroup of patients was

identified as high risk if they had at least four of the following: SCr

greater than 2.0 mg/dL during the preceding month, NYHA class

IV for the preceding 2 months, older than 65 years of age, a history of sustained ventricular tachycardia, ischemic HF etiology,

diabetes, or use of nesiritide or inotropic agents as outpatients

within the preceding 6 months. These patients at high risk experienced fewer HF exacerbations and renal adverse effects with

nesiritide infusions.

This suggestion of potential benefit and safety in outpatients

with severe HF was further explored in the Follow-Up Serial Infusions of Nesiritide for the Management of Patients With Heart

Failure (FUSION II) trial.311 This was a randomized, placebocontrolled, double-blind prospective trial of 911 subjects with

advanced HF or chronic decompensated HF. Subjects were randomly assigned to receive a 2-mcg/kg nesiritide bolus followed

by a 0.01-mcg/kg/minute infusion for 4 to 6 hours, or a matching placebo regimen, once or twice weekly for 12 weeks. Both

groups also received optimal medical therapy and device therapy.

Patients were entered only if their creatinine clearance was less

than 60 mL/minute. No outpatient IV inotropic or vasodilator

therapy was allowed during the 24-week study. After 24 weeks,

no difference was noted in the primary end point of either death

or hospitalization for cardiac or renal causes among patients in

the nesiritide and placebo groups. Significantly more (p <0.001)

drug-related adverse events occurred in those in the nesiritide

group (42.0%) compared with the placebo group (27.5%), mainly

caused by hypotension. The incidence of worsening renal function, however, was significantly lower (p = 0.046) with nesiritide

(32% in nesiritide group vs. 39% in the placebo group). There

was no evidence that changes in SCr were associated with renal

harm.

The results of these studies alleviated some concerns regarding the safety of nesiritide; however, a group of independent

cardiologists designed an “Acute Study of Clinical Effectiveness of Nesiritide in Decompensated Heart Failure” (ASCENDHF)312,313 trial that would help to definitively answer the question

of nesiritide safety and efficacy. This was a double-blind placebo

trial, which enrolled 7,141 patients with ADHF in 30 countries.

The participants were randomly assigned to receive IV bolus

nesiritide (loading dose) of 2 mcg/kg or placebo (investigator’s

discretion for bolus), followed by continuous IV infusion of

nesiritide 0.01 mcg/kg/minute or placebo for 7 days in addition

to standard therapy. The coprimary end points were to assess a

reduction in the rate of HF hospitalizations or all-cause mortality through day 30 and a significant improvement in self-assessed

dyspnea at 6 or 24 hours using a 7-point Likert scale. Compared

with placebo, nesiritide was not associated with a reduction in

30-day mortality or rehospitalization (10% vs. 9.4%; p = 0.31).

Nesiritide improved dyspnea at 6 hours (15% vs. 13.4%; p = 0.30)

480 Section 2 Cardiac and Vascular Disorders

and at 24 hours compared with placebo (30.4% vs. 27.5%;

p = 0.007), which is consistent with previous findings but did

not meet the prespecified criteria or statistical significance at 6 or

24 hours. Also, nesiritide did not worsen renal function as had

been suggested by prior meta-analysis. The proponents of nesiritide argue that even though the ASCEND-HF trial failed to show

a major benefit, it is the only vasodilator that has been well

studied. The controversy regarding the safety and survival of

prescribing nesiritide can be put to rest; however, further analysis of the ASCEND-HF trial will provide better understanding

of ADHF and patient profiles that may potentially benefit from

nesiritide. The principal take-home message is that large clinical trials should be the norm before drugs are introduced into

clinical practice.288

L.M. responded well to nitroprusside and needed no other

therapy. However, IV NTG and IV nesiritide are alternatives to

nitroprusside in patients with “wet and warm” acute HF. NTG or

nesiritide can either be a substitute for nitroprusside or combined

with nitroprusside. At the time of discharge, furosemide therapy should be continued, although the dose should be titrated

to prevent DOE and peripheral edema. The dose of lisinopril

and carvedilol should also be titrated to target or tolerated

doses.

Inotropic Agents

CASE 19-5

QUESTION 1: B.J. is a 60-year-old man who presents to the

emergency department with worsening DOE. He reports

increased SOB during the last week. His medical history

includes HTN, CAD, hyperlipidemia, and HF (EF 25%). His

medications include metoprolol succinate 100 mg every day,

enalapril 5 mg every day, furosemide 40 mg BID, aspirin

81 mg, and lovastatin 20 mg every night at bedtime. His

vital signs on admission included BP of 92/70 mm Hg, HR

of 92 beats/minute, respiratory rate of 18 breaths/minute,

and O2 saturation of 94% on room air. Laboratory values

were BUN of 20 mg/dL and SCr of 1.4 mg/dL. Physical

examination reveals pulmonary and peripheral edema. He

is given 80 mg IV furosemide, and he does not respond adequately. His SCr has increased to 1.8 mg/dL. The decision is

made to admit him to the coronary care unit where a pulmonary artery catheter (Swan–Ganz catheter) is placed, and

the following hemodynamic variables are measured and calculated: PCWP, 21 mm Hg; cardiac index, 1.8 L/minute/m2;

SVR, 580 dyne · s/cm5. Is B.J. a candidate for inotropic therapy? Why is milrinone preferred over inamrinone?

DOPAMINE AND DOBUTAMINE

The hemodynamic profile of B.J. is category IV (cold and

wet) because of hypoperfusion and congestion. According to

the ACC/AHA guidelines, IV inotropic agents (dobutamine,

dopamine, milrinone, inamrinone) are indicated in symptomatic

patients with reduced LVEF, low CO, or end-organ dysfunction

(i.e., worsening renal function) and in patients who are intolerant

to vasodilators.1 They are also recommended for patients with

cardiogenic shock or refractory symptoms, and may be used in

patients requiring perioperative support after cardiac and noncardiac surgery or for those awaiting transplantation. Dobutamine

should be the drug of choice in patients with low-output HF.

Dobutamine improves CO, decreases PCWP, and decreases total

SVR with little effect on HR or systemic arterial pressure when

compared with dopamine.314 (See Chapter 22, Shock, for further

information on dopamine and dobutamine.)

PHOSPHODIESTERASE INHIBITORS: INAMRINONE

AND MILRINONE

Although β-agonists, such as dobutamine, have been traditionally used in patients with ADHF, the phosphodiesterase

inhibitors, inamrinone and milrinone, are alternatives to the

catecholamines and vasodilators for the short-term parenteral

treatment of severe congestive failure. These agents selectively

inhibit phosphodiesterase III, the cyclic adenosine monophosphate (cAMP)-specific cardiac phosphodiesterase. They have

direct cardiac-stimulating effects, but they are not sympathomimetics or inhibitors of Na+/K+ ATPase. Enzyme inhibition

results in increased cAMP levels in myocardial cells and, thus,

enhances contractility. Their activity is not blocked byβ-blockers.

Because they are phosphodiesterase inhibitors, they also act as

vasodilators. It has been suggested that at low doses they act

more as unloading agents rather than inotropic agents; others

refute this viewpoint. Their overall hemodynamic effect probably results from a combination of positive inotropic action plus

preload and afterload reduction.

Inamrinone is no longer used because of dose-dependent

reversible thrombocytopenia (up to 20% of patients), drug fever,

liver function abnormalities, and possibly drug-induced ventricular arrhythmias. Milrinone is structurally and pharmacologically similar to inamrinone.315–317 Besides inhibiting phosphodiesterase, it also may increase calcium availability to myocardial

muscle. It has both inotropic and vasodilating properties. HR

increase and myocardial consumption may be less with milrinone than with dobutamine.318,319 The half-life of milrinone is

short (1.5–2.5 hours), with renal clearance accounting for approximately 80% to 90% of total body elimination. Although a loading

dose of 50 mcg/kg administered during 10 minutes is in the product labeling, it is rarely given. Maintenance infusions are typically

between 0.2 and 0.75 mcg/kg/minute. The infusion is adjusted

according to hemodynamic and clinical responses and should

be decreased in patients with renal insufficiency. The primary

concern with the use of milrinone is induction of ventricular

arrhythmias, reported in up to 12% of patients. Supraventricular

arrhythmias, hypotension, headache, and chest pain also have

been reported. Thrombocytopenia is rare, a distinct advantage

compared with inamrinone. Overall, milrinone has become the

drug of choice among the phosphodiesterase inhibitors.

The Outcomes of a Prospective Trial of Intravenous Milrinone

for Chronic Heart Failure (OPTIME-CHF) trial assessed the inhospital management of 951 patients with acute HF exacerbation

(NYHA class III or IV, mean LVEF 23%) but not in cardiogenic

shock.318,319 In addition to standard diuretic and ACE inhibitor

therapy, subjects were randomly assigned to receive either milrinone or placebo. The initial milrinone infusion rate was 0.5

mcg/kg/minute with no loading dose. For the primary end point

of total numbers of days hospitalized for CV causes from the

time of the start of study drug infusion to day 60, no difference

was found between active drug and placebo (mean 12.3 days with

milrinone and 12.5 days with placebo), nor was any difference

seen in the mean number of days of hospitalization during the

primary event. Death rates within 60 days were 10.3% with milrinone and 8.9% with placebo (not significant, p=0.41).318 Followup analysis categorized subjects by etiology of HF (ischemic vs.

nonischemic).319 Not unexpectedly, those with an ischemic cause

did less well with hospital rates of 13.0 days for ischemic patients

compared with 11.7 days for those without ischemia (p = 0.2).

Corresponding death rates for 60 days were 11.6% and 7.5%

(p = 0.03). Importantly, within the cohort of patients with

ischemia (n = 485), milrinone-treated patients tended to have

worse outcomes than those treated with placebo: 13.6 hospital days with milrinone versus 12.4 days with placebo. Death

occurred in 13.3% of ischemic patients on milrinone compared

481Heart Failure Chapter 19

with 10.0% of those on placebo. For the composite of patients

dying or being rehospitalized, 42% of milrinone subjects had

events compared with 36% with placebo (p = 0.01). In contrast,

nonischemic patients (n = 464) had a trend toward better outcomes with milrinone than with placebo: 10.9 hospital days for

milrinone versus 12.6 days with placebo, 7.3% deaths with milrinone compared with 7.7% with placebo, and the composite of

death or hospitalization in 28% of subjects with milrinone versus

35% with placebo. From these data, it can be concluded that the

benefits of milrinone in patients with acute exacerbations of HF

are minimal and more likely to be seen in patients with a nonischemic etiology of HF. Worse outcomes may be seen in patients

with ischemic HF. However, short-term infusions of milrinone

were not associated with excess mortality.

Few trials have compared dobutamine with milrinone in

ADHF patients. One small retrospective analysis evaluated 329

patients admitted for ADHF with an EF of less than 20% who

either received IV dobutamine or milrinone.320 Hemodynamic

response, need for additional therapies, adverse effects, length

of stay, and drug cost were evaluated. Patients in both groups

were comparable in clinical presentation. Further, both groups

had similar HR, BP, PCWP, and CO at baseline. The milrinone

group, however, had higher mean pulmonary arterial pressure.

A greater percentage of patients received dobutamine therapy

(269, 81.7%) versus milrinone therapy (60, 18.3%.). Only 19% of

patients were taking β-blockers before admission. Clinical outcomes in both groups were similar. There was no significant

difference in the in-hospital mortality rate, adverse effects, ventilator use, or length of stay. More patients in the dobutamine

group required nitroprusside to achieve optimal hemodynamic

response. A slightly better hemodynamic response occurred in

the milrinone group, which did not translate into a more beneficial short-term clinical outcome. The study concluded that both

drugs have comparable efficacy.

Other factors to consider when deciding which inotropic

agent should be used in ADHF are renal function, BP, and concomitant β-blocker use. Milrinone has a longer half-life than

dobutamine, and accumulates in cases of renal dysfunction. Milrinone is also a vasodilator, which can limit its use in patients with

hypotension. The concomitant use ofβ-blockers may antagonize

the action of dobutamine.

B.J. has responded poorly to IV furosemide, his renal function has deteriorated, and his systolic BP is low. Patients with

advanced HF, reduced BP, and normal or low SVR often will

not tolerate vasodilator therapy. Inotropic agents may be necessary to maintain circulatory function in these patients. According to guidelines, IV inotropic drugs may be considered for

patients who have symptomatic hypotension despite adequate

filling pressure, who are unresponsive to diuretics and intolerant

to vasodilators, or who have worsening renal function. Phosphodiesterase inhibitors are sometimes preferred instead of dobutamine for patients who are receiving concomitant β-blockers.

For the above-mentioned reasons, B.J. is a candidate for milrinone therapy. Patients must be on telemetry because milrinone

has the potential to cause arrhythmias. Vital signs, SCr, symptom

relief, and urine output should be monitored. Once the patient’s

hemodynamic profile improves, milrinone should be discontinued, and oral furosemide therapy can be resumed. At discharge,

the outpatient HF medications should be optimized.

Outpatient Inotropic Infusions

CASE 19-5, QUESTION 2: Are there any indications for

using repeated intermittent infusions of inotropic agents

as part of a home-care regimen?

The long-term safety and efficacy of inotropic therapy in

general is regarded with skepticism. There are few studies assessing intermittent (e.g., weekly) infusions of dobutamine or milrinone. Nearly all the data on this therapeutic approach are

from open-label and uncontrolled trials or studies that compare two inotropic agents without a placebo group.321–325 It is

unclear whether the benefit observed was the result of more

intensive patient monitoring or an actual pharmacologic benefit. It is also speculated that long-term therapy may actually be

cardiotoxic, as evidenced by an acute worsening of HF on withdrawal of the drug. The only placebo-controlled trial of intermittent infusion of dobutamine was terminated because of excess

mortality in the treatment group.322 Death occurred in 32% of

31 patients treated with dobutamine and only 14% of 29 treated

with placebo. Whether this phenomenon was caused by progression of the underlying heart disease or continued drug therapy

or was a true cardiotoxic effect remains unknown. No corresponding data exist for milrinone, although as cited previously,

a placebo-controlled trial with milrinone failed to support the

routine use of IV milrinone as an adjunct to standard therapy in

the treatment of patients hospitalized for an acute exacerbation

of chronic HF.318,319 For this reason, the ACA/AHA guidelines

explicitly indicate that intermittent infusions of dobutamine and

milrinone in the long-term treatment of HF, even in advanced

stages, should be avoided.4 Dobutamine and milrinone are sometimes administered as long-term infusions in patients with refractory HF who are awaiting transplant. The lowest dose possible

should be administered.

Ventricular Arrhythmias Complicating

Heart Failure

AMIODARONE

CASE 19-5, QUESTION 3: B.J. was stabilized during the next

several days and discharged home with furosemide 40 mg

every day, enalapril 5 mg every day, metoprolol succinate

100 mg every day, aspirin 81 mg every day, and NTG 0.4 mg

sublingual to be used as needed for chest pain. His EF was

23%. Laboratory values were all normal. ECG monitoring

during B.J.’s hospital stay showed normal sinus rhythm, but

he was having 15 to 20 asymptomatic PVCs/hour. At that

time, it was decided not to treat his arrhythmia other than

with metoprolol because he was asymptomatic. For the next

several months, he continued to have frequent PVC during

follow-up examinations in the cardiology clinic.

It has now been 5 months and he is still having up to 12 to

15 PVCs/hour. His exercise capacity is limited by SOB after

walking about a block despite having his enalapril increased

to 20 mg/day and metoprolol succinate to 200 mg/day and

adding digoxin 0.25 mg every day. The furosemide is still at

40 mg/day because he has some edema. A repeat echocardiogram shows an EF of 20%. Is an antiarrhythmic agent

indicated for B.J. at this time? What is the agent of choice,

and what dose should be given?

PVCs and other arrhythmias are a common complication of

LV dysfunction and may be present regardless of whether the

patient has had an MI. Approximately 50% to 70% of patients

with HF have episodes of nonsustained ventricular tachycardia

on ambulatory monitoring.4 This myocardial irritability may

be a result of autonomic hyperactivity or ventricular remodeling that can accompany HF. It is not clear, however, whether

these rhythm disturbances contribute to sudden death or simply reflect the underlying disease process. Recent studies suggest

482 Section 2 Cardiac and Vascular Disorders

that bradyarrhythmia or electromechanical dissociation may be

associated with sudden death in HF patients with nonischemic

cardiomyopathy.326,327 More importantly, suppression of ventricular ectopy in patients with HF has not been shown to lead to

a reduction of sudden death in clinical trials. B.J.’s PVCs were

first noted after his MI. As discussed in detail in Chapter 18,

Acute Coronary Syndrome, and Chapter 20, Cardiac Arrhythmias, neither prophylactic antiarrhythmic therapy nor treatment

of asymptomatic PVC after an MI has been proven to improve

outcome or survival. In fact, because of concerns about proarrhythmic effects of most class IA (e.g., quinidine) and class IC

(e.g., flecainide) drugs, as well as sotalol, treatment is considered

contraindicated.

It is suggested that amiodarone has value in patients with HF

with arrhythmias because it has both antiarrhythmic properties

as well as coronary vasodilating effects and α- and β-blocking

properties. Thus, it may offer a dual benefit to reduce myocardial

irritability and improve the hemodynamics of HF.

One meta-analysis reviewed 13 randomized, controlled trials

of prophylactic amiodarone in patients with either recent MI

(n = 8) or HF (n = 5).328 None of the individual trials were powered to detect a mortality reduction of less than 33%. Therefore,

moderate reductions of mortality that still may be clinically relevant would not have been identified as statistically significant.

After loading doses of 400 to 800 mg/day for 2 weeks, maintenance doses ranged from 200 to 400 mg/day. The authors concluded that prophylactic amiodarone reduces the rate of arrhythmic or sudden death in high-risk patients, and this effect results in

an overall 13% reduction in total mortality. Because this analysis

combined trials of both MI and HF patients, it is helpful to look

at two of the key HF trials.

In the Grupo de Estudio de la Sobrevida en la Insuficiecia

Cardiaca en Argentina (GESICA) study,329 516 patients with class

II to IV HF symptoms (79% class III or IV), an average EF of

20%, and frequent PVCs on cardiac monitoring were randomly

assigned to receive either standard treatment (diuretics, vasodilators, digoxin) or a fixed dose of amiodarone plus standard treatment. The dose of amiodarone was 600 mg daily for the first

2 weeks, then 300 mg/day for at least 1 year. Of 260 patients on

amiodarone, 87 (33.5%) died during follow-up compared with

106 of 256 (41.4%) receiving standard treatment, a statistically

significant difference in favor of amiodarone (p = 0.02). Similarly, the number of HF-related hospitalizations was reduced

with amiodarone. No data were presented on changes in EF, but

a trend toward more patients in the amiodarone group being

judged to have a decrease of at least one stage in NYHA class was

noted.

Somewhat different outcomes were noted by the investigators in the Veterans Administration (VA) Cooperative Survival

Trial of Antiarrhythmic Therapy in Congestive Heart Failure

(CHF-STAT) study.330,331 Entry criteria to this trial were similar to the GESICA study, with a primary indicator being more

than 10 asymptomatic PVCs per hour on 24-hour monitoring,

but without sustained ventricular tachycardia. A higher dose of

amiodarone was used, starting with 800 mg for the first 2 weeks,

then 400 mg/day for 1 year. The dose was reduced to 300 mg/day

after the first year, with the average follow-up being 45 months

(4.5 years maximum). No difference between groups for either

all-cause mortality (39% amiodarone vs. 42% placebo) or sudden cardiac death (15% amiodarone vs. 17% placebo) was found.

Similarly, 2-year survival was 69.4% with amiodarone and 70.8%

with placebo. Higher survival in the amiodarone group after

the first 2 years was a noted trend, but the number of subjects

observed for longer periods was not sufficiently large to establish significance. An encouraging finding in this study was that

EF improved more in the patients treated with amiodarone, rising from a baseline average of 24.9% to posttreatment values of

33.7%. Corresponding change in the standard treatment group

was from a baseline of 25.8% to 29.2% at follow-up. Despite the

increase in EF, symptom scores did not differ between the two

groups.

Together, these two studies still leave unclear the role of amiodarone in patients with HF with asymptomatic arrhythmias. The

encouraging finding is that amiodarone does not seem to have

a negative effect on mortality as seen with some of the other

antiarrhythmic agents. On the other hand, the two studies cited

have conflicting findings regarding value in improving survival

and functional capacity of patients. In comparing the two studies,

it has been noted that the patients in the GESICA study had more

advanced disease (79% class III or IV; average EF 20%; 55% 2-year

placebo mortality) than those in the VA study (43% class III or IV;

average EF 25%; 29% 2-year placebo mortality); more patients

had nonischemic cardiomyopathy in the GESICA study (60%)

compared with 29% in the VA study; 99% of the VA subjects

were men, whereas 19% of GESICA subjects were women; and

the dose of amiodarone was lower in the GESICA study.332 Further subgroup analysis suggests that those with more advanced

disease, nonischemic cardiomyopathy, and female sex have better outcomes. Contradicting this speculation was a trend toward

better outcomes in the small number of patients with class II

symptoms in the GESICA study.329 Other factors to consider are

the potential for significant side effects with amiodarone (see

Chapter 20, Cardiac Arrhythmias) and the risk that digoxin levels

increase after the addition of amiodarone.

The ACC/AHA guidelines do not recommend routine ambulatory ECG monitoring to detect asymptomatic ventricular arrhythmias in patients with HF, and they also recommend against treatment if such arrhythmias are inadvertently

detected.21 If symptomatic ventricular arrhythmias should arise

or there is determined to be a high risk for sudden death, one

of the following should be considered: a β-blocking drug, amiodarone, or an ICD. As discussed extensively throughout this chapter, nearly all patients with HF should have a β-blocker as part

of their regimen because these drugs reduce all-cause mortality,

not just sudden death. B.J. continued to have ectopy despite continued use of a β-blocker. Nonetheless, it is decided not to use

amiodarone because he has ischemic cardiomyopathy and is not

bothered by his arrhythmia.

IMPLANTABLE CARDIOVERTER-DEFIBRILLATOR

CASE 19-5, QUESTION 4: Is B.J. a candidate for an ICD

implantation?

Although amiodarone is the preferred antiarrhythmic agent

in patients with HF with reduced EF to prevent recurrent AF

and symptomatic ventricular arrhythmias, it has not improved

survival benefits. Ventricular arrhythmias are associated with

high frequency of SCD in patients with HF. Numerous trials

have established the role of ICDs in primary and secondary prevention of SCD. The earliest of the primary prevention trials

was the Multicenter Automatic Defibrillator Implantation Trial

(MADIT).333 This study was terminated early because of the survival benefit seen in the ICD group compared with conventional

therapy (hazard ratio, 0.46; p = 0.009). There was no evidence

that amiodarone, β-blockers, or any other antiarrhythmic therapy had a significant influence on the observed hazard ratio.

Unlike MADIT, the MADIT II334 study enrolled patients with no

documented arrhythmias but with previous MI and LVEF less

than 30%. Patients received either an ICD or conventional medical therapy. The primary end point was death from any cause.

There was a 31% relative reduction in the risk of death and an

absolute reduction of 6% in the ICD group compared with the

483Heart Failure Chapter 19

medical group. This was the first trial to show mortality benefits

of ICDs in patients with no documented history of abnormal

heart rhythms.

The most recent trial, the Sudden Cardiac Death in Heart

Failure trial (SCD-HeFT) evaluated the efficacy of amiodarone in

patients with LV dysfunction (EF≤35%).332 The patients (NYHA

class II–III) were randomly assigned to conventional therapy

or placebo, conventional therapy plus amiodarone, or conventional therapy plus ICD. Amiodarone was no better than placebo,

whereas ICD decreased mortality by 23% (p = 0.007) compared with conventional therapy. A subgroup analysis showed

that patients with class II HF had a greater drop in mortality

with ICD use than class III patients. Also, amiodarone decreased

survival in class III HF. The role of amiodarone in patients with

NYHA class III needs to be further evaluated before it is routinely

used in patients with LV dysfunction.

The 2009 ACC/AHA guidelines recommend the use of ICDs

in patients after MI with reduced LVEF and who have a history of

ventricular arrhythmias. ICDs are also recommended for primary

prevention in patients with nonischemic cardiomyopathy and

ischemic heart disease who have an LVEF of 30% or less, those

with with NYHA functional class II or III symptoms while on

optimal standard oral therapy, and patients who have reasonable

expected survival with a good functional status of 1 or more

years. Patients with ischemic heart disease should be at least

40 days after MI to receive an ICD. B.J. is currently on optimal

HF drug regimen, and his EF is 20%. According to guidelines,

B.J. would benefit from ICD implantation.

CARDIAC RESYNCHRONIZATION THERAPY

CASE 19-6

QUESTION 1: C.M., a 49-year-old woman with a history of

cardiomyopathy (EF 25%), presents to the HF clinic with

NYHA class III symptoms. She reports increased SOB, chest

pain, and fatigue. She has been optimized on drug therapy

for 3 months. Her medications include metoprolol succinate

200 mg every day, furosemide 40 mg BID, lisinopril 20 mg

every day, spironolactone 25 mg every day, and potassium

chloride 40 mEq. An ECG showed sinus rhythm at a rate of

72 beats/minute and a QRS duration of 144 milliseconds. Is

she a candidate for CRT?

Approximately one-third of the patients with advanced systolic HF exhibit intraventricular or interventricular conduction

delays that cause the ventricles to beat asynchronously.335 This

ventricular dyssynchrony, which is often seen on the ECG as a

wide QRS complex with a left bundle branch block, can lead to

deleterious effects on cardiac function. Patients may present with

reduced EF, decreased CO, and presence of NYHA class III or IV

HF symptoms. These are all associated with increased mortality.

CRT is the use of cardiac pacing to coordinate the contraction

of the left and right ventricles.78 Initial randomized trials of CRT

show reduced HF symptoms and improved exercise tolerance