Cardiogenic Shock

Shock is characterized by a state of end-organ hypoperfusion resulting in abnormal organ

homeostasis, leading to high patient morbidity and mortality. Cardiogenic shock (CS) is a clinical
syndrome characterized by systemic hypotension and hypoperfusion secondary to insufficient
cardiac output. In states of pure CS, cardiac filling pressures are elevated and cardiac output is
low. CS can lead to multisystem organ failure, manifested by oliguria, lactic acidosis, altered
mentation, and cool extremities. Most commonly, CS is the direct sequelae of an acute
myocardial infarction (MI), and acute ischemic CS carries an in-hospital mortality of greater than
50%. However, CS can also arise as an acute presentation of a cardiomyopathy of nonischemic
cause or as a severe decompensation of chronic (ischemic or nonischemic) cardiomyopathy. The
latter presentations are less common and account for only 1% of acute heart failure syndromes.
Critical care management is centered on an efficient, rapid, and organized approach to the shock
patient using a multidisciplinary care approach between intensivists, heart failure specialists,
cardiac surgery, and interventional cardiology. The tenets of therapy include restoring cardiac
output and identifying and treating one or multiple potential causative factors: hypoxia,
hypervolemia, acidosis, arrhythmias, coronary ischemia, and mechanical complications of an MI.
In certain clinical situations (eg, requirement for multiple inotropes or vasopressors, worsening
hemodynamics despite inotrope support), it may be prudent to refer the patient to the nearest
left ventricular assist device (LVAD)/transplant program for further management.

PHYSICAL FINDINGS
The importance of a focused physical examination in the management of a patient with CS
should be emphasized. It is at this critical juncture in the clinical evaluation process that
obtaining the correct information can direct the caregiver down the right diagnostic and
therapeutic pathway. A thorough assessment allows for evaluation of intravascular volume status
and adequacy of end-organ function (Box 1).

Jugular Venous Pressure and S3 Gallop

The internal jugular vein forms a direct fluid column with the right atrium and provides a
noninvasive measure of right atrial pressure. The correct method for determining jugular venous
pressure (JVP) is depicted in Fig. 1. The patient should be placed in bed at a 45 angle, which can
be confirmed by use of the ball found commonly on the side of a hospital bed. Careful
observation of internal jugular vein with the head and neck as aligned in Fig. 1 allows one to
appreciate the JVP. It is important to avoid having the patient extend their neck because this can
cause the internal jugular vein to flatten.
Once the meniscus of the JVP has been located, an estimate of the right atrial pressure can be
made by adding 5 cm to the distance of the JVP meniscus above the angle of Louis. The constant
of 5 cm is added because it represents the distance from the sternum to the right atrium. The
normal JVP is 6 to 8 cm of water, and 1 cm of water is equivalent to 0.74 mm Hg. In patients who
have a very low or high JVP, it is often necessary to lay the patient flat or elevate the head of the
bed at 90_, respectively, to appreciate the meniscus of the JVP. 5,6
The ventricular gallop, S3, or third heart sound is a relatively specific finding of heart failure in the
adult population. In the setting of heart failure, an S 3 gallop occurs because of early and rapid
ventricular filling often in a dilated ventricle. The sound can originate from either the right or the
left ventricle and leads to a right-sided or left-sided third, early diastolic heart sound (S 3). The
quality and intensity of the S3 gallop are related to the atrial pressure, ventricular compliance,
and diastolic filling rate. An S3 tends to be louder in states of volume overload and can be heard
with significant mitral regurgitation (MR) even in those with normal left ventricular function. 7 In
pregnant women and children, an S3 may also be a benign finding.
The JVP and S3 gallop are not only useful tools for diagnosing heart failure, but are also predictive
of patient outcome. The prognostic importance of these physical examination findings was
studied within the large multicenter Studies of Left Ventricular Dysfunction trial, which evaluated
the efficacy of enalapril in the treatment of systolic heart failure. 8,9 In a multivariable analysis
that included ejection fraction, age, as well as other demographic and clinical variables, the only
predictors of death or hospitalizationfor heart failure were JVP and an S 3 gallop.9 The presence of
elevated JVP in heart failure patients was associated with an increased risk of death (relative risk
5 1.52) compared with patients without an elevation of JVP. 9 A similar increased risk of death was
seen in patients with an S3 gallop (relative risk 5 1.35) compared with those patients with heart
failure and no ventricular gallop.
Determining the Cardiac Output on Examination
JVP, dependent edema, and rales are all physical examination findings associated with fluid
overload, but give little information as to the adequacy of end-organ perfusion by the myopathic
heart. A variety of physical examination findings can be used to assist in determining whether
the cardiac output is normal or low. These findings include, but are not limited to, the presence of
cool or mottled extremities with a reduced capillary refill, the presence of a pulsus alternans, a
narrow pulse pressure, hypotension, and impaired mentation. 10 Patients can be grouped into 4
profiles based on the presence of elevated filling pressures and the adequacy of cardiac output
as outlined in Fig. 2.11
These profiles guide the usage and timing of diuretics, vasodilators, and vasoactive and inotropic
medications. These hemodynamic profiles not only guide therapy but also have prognostic
implications similar to those obtained by invasive measurements of cardiac output and
pulmonary capillary wedge pressure.12,13

Physical Examination Pitfalls

A common pitfall in the evaluation of heart failure patients is the overreliance on certain physical
examination findings to determine a volume overloaded state. Although wet crackles (ie, rales)
may be appreciated in patients with acute heart failure, they may be absent on examination of
CS patients with a past history of chronic heart failure. 14 Patients with chronic heart failure tend
to exhibit rales only in very severe states of volume overload because their pulmonary
lymphatics accommodate higher pulmonary venous pressures over time. Rales may not be heard
in some individuals with chronic heart failure presenting with acute shock until left ventricular
filling pressures are very high. Peripheral edema is neither sensitive nor specific for CS. Although
lower extremity edema can be a marker of elevated right-sided filling pressures, edema can also
occur in the setting of protein calorie malnutrition or venous incompetence.

HEMODYNAMIC ASSESSMENT
The first pulmonary arterial catheterization (PAC) was performed by Lewis Dexter in 1945 and
was performed to diagnose congenital heart disease. 15 It was not until 1970 when Swan and and
colleagues16 developed the balloon-tipped catheter that widespread use of the device became
popular. Initial excitement for the device has contemporaneously been tempered by a growing
body of evidence that its routine use in managing a variety of patient groups may be
unwarranted and potentially harmful. Over the past decade, the utilization of PACs has fallen
dramatically and some would argue that many clinicians now lack adequate training in how to
place, interpret, or manage a PAC.21,22
The routine use of PACs in patients with an acute exacerbation of heart failure was studied in the
multicenter Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheterization
Effectiveness (ESCAPE) trial.23 Patients who had their heart failure managed in conjunction with a
PAC failed to demonstrate an improvement in hospital length of stay or mortality compared with
those managed without a PAC.
The PAC group had an increased incidence of adverse events largely driven by catheter-related
complications including line infection.23 It is important to note, however, that these patients were
not necessarily in CS: the mean cardiac index was 1.9 L/min/m 2 and the trial explicitly excluded
those patients receiving an intravenous
inotrope.23
Studies specifically looking at PAC use in patients after an MI showed similar results to that of the
ESCAPE trial. In an analysis of greater than 26,000 patients presenting with an acute coronary
syndrome who were enrolled in to the Global Utilization of Streptokinase and TPA for Occluded
Coronary Arteries (GUSTO) IIb and III trials, only 735 patients received a PAC. 24 Pulmonary arterial
catheters were more commonly used in the post-MI setting when patients had higher resource
utilization (eg, coronary artery bypass grafting, percutaneous coronary intervention) and/or a
more unstable clinical presentation (eg, need for mechanical ventilation, intra-aortic balloon
pumps [IABP]).24 PAC use was associated with an increased risk of adverse events including an
adjusted 6-fold increase in 30-day mortality. However, when looking at the subgroup of patients
with CS, the use of a PAC had a neutral effect on mortality.
The nonrandomized nature of this study makes the findings about the utility ofPAC placement
questionable.24
The PAC measurements also provide prognostic information in the setting of acute MI. The
Forrester criteria assess pulmonary congestion (pulmonary capillary wedge pressure greater than
18 mm Hg) and systemic hypoperfusion (cardiac index less than 2.2 L/min/m 2) by using PAC data
to categorize patients in quartiles.25 Patients who exhibit both pulmonary congestion and
systemic hypoperfusion have a 60% inhospital mortality. It is important to note that the study
was performed in the 1970s and management of acute coronary syndromes and CS has changed
dramatically since then. Nevertheless, an appreciation for the severity of the hemodynamic
derangement allows for an objective assessment of myocardial dysfunction and may prompt
evaluation for advanced therapies including mechanical circulatory support.
Although routine invasive hemodynamic monitoring in patients with an acute heart failure
exacerbation seems unwarranted, use in CS may still be clinically indicated. Potential indications
for a PAC are listed in Box 2; these indications lack quality evidence from rigorously conducted
randomized trials but continue to be supported in professional society guidelines, largely
because of expert opinion.26,27 Further details on PACs and hemodynamic monitoring are
presented in the article in this issue by Kenaan and colleagues.

ILLNESS SEVERITY CLASSIFICATION SYSTEMS

Killip Class
The most widely recognized illness classification system to characterize patients with heart
failure after an MI was developed by Killip and Kimball in 1967. 28 These physicians are credited
with not only developing the Killip Classification system for heart failure severity (Table 1) but
also creating the modern day coronary care unit. In the global registry of acute coronary events
risk score, a higher Killip classification at time of hospital presentation was associated with a 2-
fold increased risk of death per increase in Killip class. 29,30 Similar prognostication potential is
also seen whenthe Killip classification system is applied to patients with nonST-elevation acute
coronary syndromes.31

APACHE II and SOFA Score

Although no well-validated tools have been developed for the explicit purpose of prognosticating
outcome in patients presenting with CS, various risk scores derived for predicting outcome in a
wide breadth of patients admitted to an intensive care unit (ICU) have been extrapolated to
those in CS. The Acute Physiology and Chronic Health Evaluation II (APACHE II) was developed to
prognosticate outcomes in patients presenting to the ICU with various unstable medical
maladies. The 12-item model calculates a risk score ranging from 0 to 71, taking into account
patient age, chronic illnesses/comorbidities, physiologic measurements, end-organ function, the
Glasgow coma scale, and postoperative state. Generally lower scores are assigned to patients in
the postoperative state and higher overall scores predict a higher in-hospital mortality. For those
patients with a score greater than 25, in-hospital mortality exceeds 50%. Although the initial
APACHE II validation cohort comprised a limited number of patients with cardiovascular disease,
subsequent studies applying the APACHE II score to patients presenting in CS have demonstrated
good predictive power.3335
In a study of more than 6000 patients presenting with an acute MI in Spain, each unit increase in
APACHE II score was associated with a 16% increase in mortality. 34
The Sequential Organ Failure Assessment (SOFA) score, originally devised for describing
complications in those with multisystem dysfunction due to sepsis, has also been used to predict
mortality in patients with multisystem organ failure due to cardiac causes. The SOFA score
comprises markers of renal function (serum creatinine or urine output), hepatic function (serum
bilirubin), hemodynamic stability (mean arterial pressure or use of vasopressors), neurologic
function (Glasgow coma scale), hematologic derangement (platelet count), and respiratory
stability (the ratio of PaO2/FiO2). In a retrospective analysis of 726 acute MI patients, each unit
increase in SOFA score was associated with a 1.3-fold increase in mortality. 37 The score offered
reasonable risk discrimination with an area under the receiver operating characteristic curve of
0.79.
Although both the APACHE II and the SOFA models were not derived from a CS cohort, the
leading cause of death in these patients is multisystem organ failure, which accounts for the
prognostication of both models. The models do not predict long-term outcome after hospital
discharge or death due to arrhythmias. Large studies comparing the APACHE II, SOFA, and the
(updated) APACHE III scoring systems in the patients in CS are lacking.

INTERMACS Profiles
As durable ventricular assist devices become increasingly common in the management of
patients with refractory end-stage heart failure, a unique patient registry that tracks clinical
outcomes in LVAD recipients has been devised: the Interagency Registry for Mechanically
Assisted Circulatory Support (INTERMACS).38 The registry is supported by the National Institutes
of Health, Food and Drug Administration, Centers for Medicaid and Medicare Services, industry,
and the individual institutions that participate. From the group who developed the registry, there
has also been a parallel development of a heart failure classification scheme termed INTERMACS
profiles.
INTERMACS profiles (Table 2) are more appropriate than the New York Heart Association (NYHA)
classification scheme or the American College of Cardiology Foundation/American Heart
Association (ACC/AHA) heart failure stages (AD) for categorizing disease severity and predicting
outcome in patients who have end-stage heart failure (ie, NYHA class IIIBIV symptoms and stage
D status).39,40 The profiles implicate a certain clinical course that provides prognostic information
to the health team that can assist in clinical decision-making. 41 Patients in profile 1 are termed
Crash and Burn and are in a state of severe end-organ malperfusion caused by CS. Patients in
profile 2 (also known as sliding on inotropes) have evidence of cardiac insufficiency (eg,
worsening renal function) despite inotrope dependence, and patients in profile 3 are clinically
stable, but are dependent on inotrope therapy.
Although profile 3 patients are the most stable of the described scenarios, inotrope
dependence is associated with a greater than 50% mortality at 1 year. 42 Even those who receive
intravenous inotropes for heart failure support who are not deemed inotrope dependent at
hospital discharge have higher morbidity and mortality than those with heart failure who have
never received inotropes.43

MECHANICAL COMPLICATIONS
Patients with any type of mechanical complication post-MI carry a higher mortality when
compared with patients with CS due to left ventricular dysfunction alone. 44 Previously, it was
thought that these complications occurred at a predictable time course after an MI and was
largely driven by the extent of tissue necrosis and timing of myocardial fibrosis. 45 As
revascularization strategies have improved for MI, the modern day incidence of mechanical
complications has dropped to less than 1% and most events occur within the first 24 hours of
presentation.46 Detection of a mechanical complication necessitates careful clinical attention to
patients signs and symptoms coupled with prompt echocardiographic evaluation to confirm the
diagnosis.

Right Ventricular Infarction

One of the most widely recognized, but difficult to diagnose, complications of an inferior MI is a
right ventricular infarction (RVI). The marginal branches that supply the right ventricle with blood
typically originate from the right coronary artery, the culprit vessel in most inferior MIs. RVI
complicating an inferior MI occurs at a rate of approximately 30% to 50% and accurate diagnosis
is often difficult.47 A right-sided V4 electrocardiogram lead has a sensitivity and specificity of 88%
and 78%, respectively, for the diagnosis.
Overall RVI is responsible for the development of CS in only 5% of CS cases,but carries a high
mortality.48 Patients with RVI also have a 3-fold higher risk of ventricular arrhythmias and
atrioventricular node block compared with inferior MI patients without right ventricular
involvement.49
The mainstay of therapy in RVI has been to maintain right ventricular preload by avoidance of
nitrates and diuretics. Other strategies for management include adequate saline hydration such
that the central venous pressure (ie, right atrial pressure) is above 10 mm Hg. In those patients
with significant hypotension and bradycardia, insertion of a temporary pacemaker and/or
initiation of inotrope support may be indicated to maintain a higher heart rate. This strategy of
promoting relative tachycardia seems counterintuitive in an acute MI but is often necessary to
maintain adequate left ventricular filling. Of note, the right ventricle is extremely resilient and
often shows dramatic recovery on both clinical and echocardiographic follow-up, indicating the
underlying pathophysiology of right ventricular dysfunction may represent stunning more so than
myocardial necrosis.50

Acute MR
Acute MR after MI is associated with a poor survival. 51 Risk factors for the development of MR
after MI include older age, female sex, and inferior or posterior infarction. 52
In a study of 773 patients presenting with an acute MI, mild MR occurred in 38% of subjects and
an additional 12% had moderate or severe MR.51 Event-free survival at 5 years was 84%, 74%,
and 35% for those with no, mild, and moderate or severe MR, respectively. The Should We
Emergently Revascularize Occluded Coronaries for Cardiogenic Shock (SHOCK) trial studied 1190
patients presenting in CS due to an acute MI and found that severe MR complicated 8% of patient
courses.52 Despite a mean left ventricular ejection fraction of 38%, in-hospital mortality was 55%
in those with severe MR.
Mild or moderate MR during acute ischemia is often transient and resolves after restoration of
blood flow, unlike acute papillary muscle rupture, which is lifethreatening.
Acute papillary muscle rupture occurs in about 7% of CS patients and affords about 5% of the
mortality in patients presenting with an acute MI.2,53 The occurrence of papillary muscle rupture
has to do with the location of coronary occlusion and the time to reperfusion. There are 2
papillary muscles that attach to the mitral valve leaflets via chordae tendineae: the anterolateral
and posteromedial papillary muscles. The anterolateral papillary muscle receives dual blood
supply from the left anterior descending artery and marginal branches from the left circumflex
artery, whereas the posteromedial papillary muscle has a singular blood supply from the
posterior descending artery alone. Because of the blood supply pattern, the posteromedial
papillary muscle is much more likely to rupture and this complication can be seen in the setting
of an inferior infarction.54,55 Rupture of the muscle can be either partial or complete with the
clinical severity corresponding directly to the degree of muscle rupture (Fig. 3).56 Older literature
cited an onset of rupture of 3 to 7 days post-MI, but in the contemporary era of rapid reperfusion,
the time clock for rupture has moved earlier (median time of 13 hours in the SHOCK trial) 52 and is
likely caused by reperfusion injury (inflammation) in the territory of the injured papillary muscle.
Clinically, the presentation of acute MR secondary to papillary muscle rupture is often sudden
with patients developing flash pulmonary edema (Fig. 4) and rapid hemodynamic instability.
Because of rapid equalization of pressures between the left atrium and ventricle, a murmur may
be absent. Therapy is focused on prompt clinical recognition, urgent echocardiographic
visualization, afterload reduction with vasoactive medications, and/or an IABP followed by
emergent surgical correction.5759

Ventricular Septal Defects

Unlike papillary muscle rupture, ventricular septal defects typically occur in the setting of an
anterior MI.60 The incidence is quite uncommon, occurring in only 0.2% of patients in the current
reperfusion era.61 The mortality with medical management of acute septal defect is greater than
50%, and survival at 30 days is 71% to 100% in those with rapidly recognized and surgically
corrected defects.61,62 As percutaneous treatment of structural heart disease continues to
develop, use of a septal occlusion device has shown some promise as a potential therapy for MI-
associated ventricular septal defects.63,64 Use of a septal occlusion device may be particularly
attractive in those patients who are deemed nonoperative candidates due to other medical
comorbidities.
Until a head-to-head study or large retrospective analysis comparing septal occlusion to surgery
is completed, surgery continues to be the gold standard for treatment of ventricular septal
defects in the setting of an MI (see Fig. 3).

Free Wall Rupture

Ventricular rupture of the free wall presents dramatically with electromechanical dissociation and
pericardial tamponade (see Fig. 3). Free wall rupture occurs in less than3% of patients with an
acute MI, but accounts for more than 10% of the mortality; it is a common finding at autopsy in
both out-of-hospital and in-hospital acute MI deaths. 65
Acute free wall rupture only occurs in patients with a transmural MI with a median time to onset
of 5 days post-MI. The complication is more likely to occur in patients who are older, who are
female gender, who have had anterior ST-elevation myocardial infarctions (STEMI), and for those
in whom there is a delay in coronary revascularization.66
Management is focused on hemodynamic support with fluids and vasoactive medications,
bedside pericardiocentesis, followed by prompt surgical correction.

REVASCULARIZATION
The beneficial effects of revascularization of MI patients presenting in CS were established with
the landmark SHOCK trial.67 In patients presenting with CS as a complication of their MI, early
revascularization with primary angioplasty and/or coronary artery bypass grafting was associated
with a nonsignificant reduction in the primary endpoint of mortality at 30 days when compared
with medical therapy alone. There was, however, a significant reduction in the prespecified
secondary endpoints of 6- month and 1-year mortality, with an absolute reduction of mortality by
13%.67,68 In the parallel SHOCK registry, the benefits of early revascularization were similarly
noted.2 At the time of the SHOCK trial in the late 1990s, early revascularization time was defined
as occurring within 6 hours of presentation and only 36% of the revascularization patients
received coronary stenting. In the current door-to-balloon era of revascularization, where most
MI patients receive prompt revascularization with primary percutaneous coronary intervention,
stent placement, and aggressive adjunctive medical therapy (ie, dual antiplatelet therapy,
statins, anticoagulants, b-blockers, angiotensin converting enzyme inhibitors (ACE-I) and
angiotensin II receptor blockers (ARBs), and aldosterone blockade), the rates of CS have dropped
to approximately 5% for patients presenting with an STEMI.69 Furthermore, general trends have
shown a decreased incidence in the overall rate of STEMI over time, dropping from 47% in 1999
to 23% by 2008, which has also contributed to the reduced rate of CS over time. 70 The mortality
from CS has also dropped from 60% in 1995 to 48% in 2004. 71
One controversial finding in the original SHOCK trial was the patient group aged 75 years or older
had no additional mortality benefit with revascularization when compared with their younger
counterparts.68 However, conclusions from this subgroup analysis should be interpreted with
caution because only one-sixth of the overall randomized trial population was aged 75 or older.
Results of subsequent clinical registries have shown that the benefit of revascularization extends
to those over the age of 75 and even the oldest old, aged 85 years or older. 2,72 Reflecting this
clinical data, the most recent ACC/AHA guidelines no longer use age to differentiate which
patients will benefit from revascularization.73 If primary angioplasty is not available, thrombolytic
therapy should be initiated, although the results are generally less favorable.

MEDICAL MANAGEMENT
Medical management of patients in CS should focus on improving cardiac output and addressing
complications of CS (eg, electrolyte disturbances, hypoxia) that may amplify the effects of shock.
Options for medical management include intravenous inotrope and vasopressor support, and
patients in severe CS often need both to maintain organ perfusion. Patients with low-grade
CS/insufficiency who are not requiring vasopressor support may (paradoxically) gain benefit from
the cautious addition of vasodilators.
Inotrope Support
The routine use of inotrope support in heart failure for short-term or long-term support is clearly
linked to an increased mortality.43,75 The routine use of these agents in management of most
heart failure syndromes is inappropriate and should be discouraged, but they have an important
role in maintaining systemic perfusion and restoring end-organ function for patients in CS. 26 Both
dobutamine and milrinone increase the inotropy of the failing heart to improve cardiac output.
Dobutamine stimulates both b1 receptors and b2 receptors, triggering the G-protein adenylate
cyclase cascade that leads to increased cyclic AMP production. Although dobutamine is mainly a
b1 agonist, stimulation of peripheral b2 receptors can lead to a drop in blood pressure noted on
medication initiation. Typical doses of dobutamine range from 2.5 to 20 mg/kg/min. In rare cases,
patients may develop an allergic reaction to dobutamine, which is manifested as acute,
unexplained, renal failure and eosinophilia in both urine and blood smears and (often) evidence
of eosinophilic infiltration on myocardial biopsy. Discontinuation of the agent is required and a
reintroduction of the medication in the future should be done with caution as recurrence is known
to occur.
Milrinone is a selective phosphodiesterase-3 inhibitor that increases intramyocyte cyclic AMP
levels leading to increased intracellular calcium for myofilament binding.
The net result is a vasodilatory effect in the pulmonary and systemic circulations and increased
inotropy within the heart without significant chronotropic alterations. 76
Typical milrinone doses are 0.125 to 0.75 mg/kg/min. Because of a long half-life (2.5 hours), the
agent takes about 7 hours before peak effects can be seen. The long time to drug onset may be
offset with an intravenous bolus load (50 mg/kg/min), but this practice is strongly discouraged
(especially in unstable patients) due to the increased risk for acute hypotension. Active and
inactive metabolites of milrinone are renally cleared and dose adjustments should be made in
patients with low glomerular filtration rates.
Both dobutamine and milrinone are associated with an increased risk of atrial and ventricular
arrhythmias and systemic hypotension. Dobutamine has a shorter halflife, which is associated
with an earlier onset of action and elimination from the body should ectopy or hypotension
develop. Because of milrinones long half-live, it is the preferred agent for outpatient parenteral
therapy and is a more potent pulmonary arterial vasodilator. 77 Clinical outcomes are similar and
the choice of agent is generally determined by clinician preference, institutional availability, and
potential need to transition to outpatient parenteral therapy.78
One other intravenous inotropic agent is Levosimendan. This drug binds to troponin C and
sensitizes the myofilament to calcium. When compared headto- head with dobutamine, there was
no beneficial effect on clinical outcomes at 180 days and this drug remains unapproved for
clinical use within the United States.

Vasodilators
Although the use of a vasodilator in patients with critical CS is contraindicated, they can be
initiated with caution in patients with low-grade shock. Intravenous vasodilators include
nitroglycerin, nitroprusside, and nesiritide. Nitroglycerin is a strong venodilator that is effective in
reducing preload and in vasodilating the coronary vasculature. Unfortunately, tachyphylaxis
requiring dose escalation is common, limiting its clinical application to mainly those patients with
refractory angina. Nitroprusside vasodilates the arterial and venous vasculature by means of the
guanyl cyclase pathway. This agent is commonly used in acute heart failure syndromes in
patients without evidence of severe shock to reduce systemic and pulmonary afterload.
In selected patient with lower grades of CS stabilized with inotropes, the addition of nitroprusside
may lead to a reduction in left and right ventricular afterloads, leading to improved left-sided and
right-sided stroke volumes. Paradoxically, because of the benefits in cardiac output, blood
pressures can even increase with nitroprusside therapy. In head-to-head comparisons with
inotropic agents, nitroprusside has been shown to reduce the systemic and pulmonary vascular
resistance, pulmonary capillary wedge pressure and improve cardiac output as effectively as an
inotrope. The very short half-life of nitroprusside compared with other intravenous vasodilators
makes it particularly attractive for ICU management of those with cardiac insufficiency.
Nitroprusside should be started at low doses (0.5 mg/kg/min) with an arterial line in place.
The dose may be titrated by 0.5 mg/kg/min increments while maintaining a goal blood pressure.
It is important to monitor patients for signs and symptoms of cyanide toxicity. Patients with
cyanide toxicity may present with confusion, nausea, vomiting, or hyperreflexia and laboratory
test results may demonstrate new or worsening lactic acidosis. Monitoring of serum thiocyanate
levels is useful if provided by an in-hospital laboratory in a timely fashion. Toxicity is more
common in patients with renal dysfunction and with prolonged administration.
The last class of intravenous vasodilators used for patients in CS includes nesiritide.
Nesiritide is a recombinant B-type natriuretic peptide that is an arterial and venous vasodilator
and has natriuretic peptide properties. Initial studies of this drug showed not only a favorable
hemodynamic profile but also improved short-term mortality. 83
Later pooled analyses showed worsening renal function and higher short-term mortality in
patients receiving nesiritide, curbing a high initial enthusiasm for the medication.
Nevertheless, the drug remains a useful adjunctive agent for managing patients with CS.

Vasopressors
For those patients with profound hypotension, use of vasopressors is often required to maintain
adequate blood pressure and organ homeostasis. Dopamine has classically been used in the
management of heart failure patients who are suffering from acute hypotension because this
medication has been shown to vasodilate the renal vasculature.
Despite these assumed beneficial effects in heart failure patients, dopamine appears to offer a
less favorable short-term mortality when compared with norepinephrine.
In 280 patients with CS managed with vasopressor support, dopamine was associated with
increased tachyarrhythmias and mortality compared with norepinephrine.
Given the high mortality for any patient receiving vasopressor therapy, the focus of treatment
should not be on the specific agent, but rather on the restoration of normal cardiac output and
resumption of normal organ homeostasis. In appropriate patients, use of temporary mechanical
support should supersede addition or further titration of vasopressors.

MECHANICAL CIRCULATORY SUPPORT

Temporary circulatory support is a promising option for management of patients in CS.

Intra-Aortic Balloon Pump

In the landmark SHOCK trial, more than 86% of patients with CS received IABP support.
In the subsequent SHOCK registry, the use of an IABP with or without thrombolytics was
associated with a significant reduction in in-hospital mortality from 72% to 50%. 88 Despite the
widespread use of IABP counterpulsation to manage patients with CS, the data supporting
favorable outcomes are quite limited, particularly in the setting of early revascularization. A
recent meta-analysis that evaluated both randomized clinical trials and observational cohort
studies of STEMI patients with CS showed an increased mortality for those patients receiving an
IABP at the time of primary percutaneous coronary intervention.89 In CS patients treated with
fibrinolytics, mortality was reduced with IABP support. 89
Two recent randomized clinical trials have been undertaken to evaluate the efficacy of IABPs for
patients in CS: the IABP-SHOCK I and II trials. The initial IABP-SHOCK trial was a small, single-
center study that randomized 45 patients with MI and CS. 90
The primary endpoint was a reduction in the APACHE II score at 4 days. There was a reduction in
APACHE II scores for both patient groups, suggesting no added benefit of IABP therapy. 90 To
confirm these results, the IABP-SHOCK II trial was conducted in 600 patients with CS complicating
an acute MI.91 Across all analyzed outcomesincluding adverse eventsthere was no difference
found between groups.91 In the most recent revision of the ACC/AHA STEMI guidelines, the use of
an IABP for patients with CS has been downgraded from a class I to a class IIa
recommendation.73,92

Percutaneous Ventricular Assist Devices

As mechanical circulatory support continues to grow with improvements in LVAD technology, a
parallel growth is occurring within percutaneous ventricular assist devices.
As the technology continues to evolve, a durable entirely percutaneous implantable ventricular
assist device is not unfathomable. Some of the more commonly used percutaneous devices
include the TandemHeart (Cardiac Assist Inc, Pittsburgh, PA, USA), Impella 2.5, Impella CP
(Abiomed Inc, Danvers, MA, USA), and peripheral extracorporeal membrane oxygenation (ECMO)
(Fig. 5). The Impella 2.5 and CP are placed across the aortic valve using a transcatheter
approach. Both devices have an axial-flow rotor that withdraws blood from the left ventricle via a
pigtail and ejects blood into the ascending aorta directly above the coronary ostia. The Impella
2.5 provides up to 2.5 L of flow, whereas the CP can provide 3.5 L of flow. Hemolysis and pigtail
migration are not infrequent complications.
The TandemHeart device withdraws oxygenated blood from an inflow cannula that is placed via
the femoral vein in the left atrium through a transseptal puncture. Oxygenated blood returns to a
pump, which sits outside the patient and is then returned via a cannula placed within the femoral
artery. The TandemHeart device is capable of providing 4 to 5 L of flow depending on the size of
the return cannula used. Complications with this device include catheter dislodgment, hemolysis,
bleeding, and limb ischemia, with limb ischemia being alleviated with the use of antegrade
perfusion catheter.

Well-powered, randomized studies of percutaneous support devices in CS are lacking.

Two small randomized studies have been conducted comparing the use of the TandemHeart
versus IABP in CS patients.93,94 Both studies showed an improvement in hemodynamic
parameters, such as cardiac index or pulmonary capillary wedge pressure, but showed no
significant difference in mortality.93,94 Each study onlyincluded approximately 40 patients and
were underpowered to assess mortality.
Another randomized study compared the Impella 2.5 device with an IABP in 25 CS patients and
found similar improvement in hemodynamics, but again no improvement in clinical outcomes. 95 A
recent meta-analysis of all 3 trials showed no improvement in short-term mortality when these
devices are used in patients with CS.96 Importantly, these studies have failed to look at the
efficacy of the percutaneous ventricular assist device when used as a bridge-to-bridge strategy
for patients who go on to receive a durable LVAD. There is some data that suggest these devices
may be effective in improving outcomes after a durable LVAD is implanted (eg, right ventricular
failure, renal failure, and operative mortality).97
The artificial heart-lung machine was developed in 1937 by John Heysham Gibbon Jr, MD to allow
performance of open heart surgery. The initial experience was complicated by hemolysis,
thrombocytopenia, and hemorrhage due to direct contact of the blood and gases used for
oxygenation. The technology was improved through the use of a membrane to separate the gas
from blood, and after years of work, the use of extracorporeal membrane oxygenation to support
adults and children with respiratory failure successfully became reported. Because of decades of
experience and a relatively inexpensive cost compared with newer percutaneous devices, ECMO
continues to be the device of choice for many institutions in patients with critical CS. It has been
used successfully as a bridge-to-bridge device to mitigate the risks of putting in a durable
LVAD.103,104 There are, however, some potential advantages of a percutaneous ventricular assist
device over ECMO: (1) the ability to decompress the cardiac chambers; (2) a reduction in wall
stress and oxygen consumption should myocardial recovery be an intent; (3) normalization of
hemodynamics (central venous and pulmonary capillary wedge pressures); and (4)
physiologically mimicking a durable ventricular assist device by allowing one a guess-estimate
of the expected response to a more durable LVAD. Because of the perceived beneficial effects of
ventricular decompression, some centers have begun to use the Impella 2.5 device with ECMO
and others decompress the LV using the TandemHeart.105 As reviewed later in this issue,
mortality following initiation of ECMO in patients with cardiac arrest and CS remains high.
Survival with the use of any percutaneous support is best when instituted before a patient is
crashing and burning.
Newer generations of percutaneous ventricular assist devices promise to increase the cardiac
output without the requirement for a larger French-size access sheath or the need for a surgical
cut down. Some upcoming devices in order of potential clinical availability include the following:
the Reitan Catheter Pump (Kiwimed, London, UK), iVAC 3L PVAD (PulseCath BV, Amsterdam, the
Netherlands), and the Percutaneous Heart Pump (Thoratec Corp, Pleasanton, CA, USA).

PERSPECTIVE FOR THE FUTURE

The improvements in clinical outcomes following the onset of CS have largely been driven by
early revascularization. Adjunctive therapies such as inotropes, IABPs, and PACs have not been
shown to improve outcomes systematically. Clinically, we exist at a plateau, where despite
continued improvements in adjunctive pharmacology and revascularization techniques, CS
mortality remains high. With newer percutaneous support devices, the ability to rest or recover
the heart until the stunned myocardium recruits or until a durable LVAD can be placed is a hope.
The field will only move forward, however, with increased randomized clinical trials such as the
SHOCK trialwhich is now more than a decade oldseemingly the only influential trial of CS
patients.