Hospital Practice: Volume 38: No.1

Nitasha Sarswat, MD And Steven M. Hollenberg, MD

Abstract: The syndrome of cardiogenic shock (CS) comprises a constellation of symptoms and signs that define a subset of patients with inadequate tissue perfusion secondary to myocardial dysfunction. Careful attention to and rapid identification of patients at risk for the development of CS and those with impending CS by both hospitalists and subspecialists will help to implement the time-sensitive therapy that it requires. Physicians should gain a familiarity with the underlying pathophysiology of CS and available diagnostic tools as well as the importance of vasopressor therapy, inotropic therapy, rapid reperfusion therapy, and mechanical support.

Keywords: cardiogenic shock; myocardial infarction; echocardiography; right heart catheterization; vassopressor therapy; inotropic therapy; intraaortic balloon counterpulsation; percutaneous coronary intervention; left ventricular assist device

Cardiogenic shock (CS), the syndrome that ensues when the heart is unable to deliver enough blood to maintain adequate tissue perfusion, is a very common reason for intensive care unit admission and is one of the most challenging emergencies for the practicing intensivist. The diagnosis portends an in-hospital mortality rate of close to 60% for all age groups.¹ Accurate and rapid identification of CS as a medical emergency, with expeditious implementation of appropriate therapy, can lead to improved outcomes.

Cardiogenic shock is a distinct subset of shock and is defined as inadequate tissue perfusion resulting from cardiac dysfunction. Clinically, CS is defined by myocardial dysfunction and tissue hypoxia in the presence of adequate intravascular volume; hypoperfusion is often manifested by altered mental state, cool/mottled extremities, and oliguria. Hemodynamically, CS can be characterized by the following parameters: systolic blood pressure < 90 mm Hg for ≥ 1 hour that is: not responsive to fluid administration alone; secondary to cardiac dysfunction; and associated with signs of hypoperfusion or a cardiac index < 2.2 L/min/m² in the setting of pulmonary capillary wedge pressure (PCWP) > 18 mm Hg.²

Risk factors for the development of CS include diabetes, hypertension, known multivessel coronary artery disease, previous myocardial infarction (MI), prior angina, peripheral vascular disease, and advanced age.³ Additional clinical risk factors include anterior wall MI, ST-segment elevation myocardial infarction (STEMI), and the presence of left bundle branch block on electrocardiogram (EKG). Cardiogenic shock seems to occur with higher prevalence in women than men; this pattern may stem from a higher rate of mechanical complications in women.

The most common cause of CS is left ventricular (LV) pump failure secondary to MI. Cardiogenic shock occurs in approximately 8% of patients with STEMIs and about 2% of the time in patients with non–ST-segment elevation myocardial infarctions (NSTEMIs), although these numbers seem to be decreasing as percutaneous coronary interventions (PCIs) become more widespread as initial treatment for acute MI.⁴ Cardiogenic shock most commonly results from extensive infarction of the anterior wall, although a smaller infarction in a patient with previously compromised LV function may also precipitate shock. Shock that has a delayed onset may result from infarct extension, reocclusion of a previously patent infarct artery, or decompensation of myocardial function in the noninfarct zone due to metabolic abnormalities. In the Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO) and Should We Emergently Revascularize Occluded Coronaries for Cardiogenic Shock? (SHOCK) registries, 89% and 75% of patients, respectively, developed CS after admission but within the first 24 hours.^1,5

Mechanical complications of infarction, such as rupture of the ventricular free wall leading to ventricular septal defects, papillary septal ruptures, and LV free wall ruptures, should always be considered in patients with CS, particularly if the clinical degree of shock is disproportionate to the size of an infarct.⁶ Higher mortality rates are seen with mechanical complications.¹

Other conditions that should be included in the differential diagnosis of CS etiology include cardiac arrhythmias (particularly ventricular tachyarrhythmias), end-stage cardiomyopathy, myocarditis, myocardial contusion, prolonged cardiopulmonary bypass, and septic shock with myocardial depression. Valvular disease (such as acute mitral regurgitation secondary to papillary muscle dysfunction), medication overdose from β-Blockers or calcium channels blockers, and hemodynamic compromise from right ventricular (RV) infarction should also be considered. Mortality rates with RV infarction may be slightly lower than LV failure, but are still substantial.^7,8

It is essential to note that an important etiology of CS is iatrogenic shock. In most cases of CS, the diagnosis is not often made until the patient has been triaged and admitted to an inpatient setting. Given that the number one cause of CS is MI, many patients receive early β-Blockade and angiotensin-converting enzyme (ACE) inhibition, which may profoundly impact the hemodynamics. It is therefore imperative that physicians are able to recognize patients with a high risk for CS and those with potential impending shock. Variables that are independently associated with excess risk of CS include older age, female sex, higher Killip class, later time from symptom onset, definite EKG abnormalities, lower blood pressure, higher heart rate, and previous hypertension.^9,10

Cardiac dysfunction in patients with CS is usually initiated by MI or ischemia. The myocardial dysfunction resulting from ischemia worsens the ischemia, creating a downward spiral. Once a critical mass of ischemic or necrotic LV myocardium (usually ~40%) loses pumping capability, stroke volume and cardiac output can begin to decrease significantly. Myocardial perfusion, which depends on the pressure gradient between the coronary arterial system and the left ventricle, and the duration of diastole, is compromised by hypotension and tachycardia, exacerbating ischemia (Figure 1). The increased ventricular diastolic pressures caused by pump failure further reduce coronary perfusion pressure, and the additional wall stress elevates myocardial oxygen requirements, worsening ischemia. Decreased cardiac output also compromises systemic perfusion, which can lead to lactic acidosis and compromise systolic performance.²

Compensatory mechanisms activated when cardiac output is reduced include sympathetic stimulation to increase heart rate and contractility, and renal fluid retention to increase preload. These compensatory mechanisms may become maladaptive and can create a vicious cycle that further worsens systolic dysfunction. Vasoconstriction to maintain blood pressure increases myocardial afterload, impairing cardiac performance and increasing myocardial oxygen demand. Myocardial ischemia increases myocardial stiffness, increasing LV end-diastolic pressure and, thus, myocardial wall stress at a given end-diastolic volume. The increased LV stiffness limits diastolic filling and may result in pulmonary congestion, causing hypoxemia and worsening the imbalance of oxygen delivery and oxygen demand in the myocardium, resulting in further ischemia and myocardial dysfunction. The interruption of this cycle of myocardial dysfunction and ischemia forms the basis for CS therapeutic regimens.²

Recent data suggest that not all patients fit into this classic paradigm. In the SHOCK trial, the average systemic vascular resistance (SVR) was not elevated, and the range of values was wide, suggesting that compensatory vasoconstriction is not universal. Some patients had fever and elevated white blood cell counts along with decreased SVR, suggesting a systemic inflammatory response syndrome.¹¹ This has led to an expansion of the classic paradigm to include the possibility that inflammatory responses contribute to vasodilation and myocardial stunning, leading clinically to persistence of shock (Figure 2).¹² Supporting this notion is the fact that the mean ejection fraction in the SHOCK trial decreased only moderately (30%), suggesting that mechanisms other than pump failure were operative.^12,13 Immune activation appears to be common in a number of different forms of shock. Activation of inducible nitric oxide synthase with production of nitric oxide and peroxynitrate has been proposed as one potential mechanism.

A key to understanding the pathophysiology and treatment of CS is to realize that areas of nonfunctional but viable myocardium can also cause or contribute to the development of CS after MI. This reversible dysfunction can be described in 2 main categories: stunning and hibernation.

Myocardial stunning represents postischemic dysfunction that persists despite restoration of normal blood flow; myocardial performance eventually recovers completely.¹⁴ Direct evidence of myocardial stunning in humans has recently been obtained by demonstrating normal perfusion using positron emission tomography scanning and ¹³N-ammonia in patients with persistent wall motion abnormalities after angioplasty for acute coronary syndromes.¹⁵ The pathogenesis of stunning has not been conclusively established but appears to involve a combination of oxidative stress, perturbation of calcium homeostasis, and decreased myofilament responsiveness to calcium. The intensity of stunning is determined primarily by the severity of the antecedent ischemic insult.¹⁴

Myocardial hibernation can be seen as an adaptive response in which segments with severely reduced coronary blood flow reduce their contractile function to restore equilibrium between flow and function, minimizing the potential for ischemia or necrosis. Function in such segments can be normalized by improving blood flow. Repetitive episodes of myocardial stunning can coexist with or mimic myocardial hibernation.^16,17

Consideration of myocardial stunning and hibernation is important because of their therapeutic implications. Both stunned and hibernating myocardium retain inotropic reserve and can respond to catecholamines. Function of hibernating myocardium can improve with revascularization and function of stunned myocardium can improve with time.¹⁴ The notion that some myocardial tissue may recover function has underscored the importance of expeditious initiation of supportive measures, including both medications and intraaortic balloon counterpulsation, to maintain blood pressure and cardiac output in patients with CS. The presence of reversible myocardial dysfunction also has prognostic implications, supported by data from the SHOCK trial, in which most survivors had only class I or class II heart failure.¹²

A thorough workup to diagnose CS includes history, physical examination, laboratory tests, including an arterial blood gas and cardiac enzymes, complete blood count, a metabolic profile as well as EKG, echocardiography, a chest radiograph, and consideration of a right heart catheter if the diagnosis is uncertain (Figure 3).

An elevated white blood cell count is a common finding given a likely inflammatory component to CS. In the case of hypoperfusion, one should expect an elevated blood urea nitrogen and creatinine, an arterial blood gas consistent with metabolic acidosis, and an elevated lactate level. Elevated liver enzymes may also be seen with hypoperfusion secondary to a “shock-liver” profile.

The physical examination should be geared toward the evaluation of congestion and systemic perfusion to act as a guide to the patient’s hemodynamic profile. An assessment of whether the patient is “wet” or “dry” and “cold” or “warm” is integral to patient management.¹⁸

Signs of congestion, indicating that the patient is “wet,” include jugular venous distension, ascites, rales, and peripheral edema. An assessment of whether the patient is “cold” or “warm” is an indication of systemic perfusion. Evidence of poor perfusion includes narrow pulse pressure, altered mental status, low serum sodium, cool or mottled extremities, renal insufficiency, and hypotension with the use of an ACE inhibitor. The majority of CS patients will be “wet” and “cold” upon presentation.¹⁸

In the setting of an acute MI, a patient may be classified according to the Killip classification, which categorizes patients by hemodynamic status and has been shown to be prognostic of mortality for patients with STEMI.¹⁹ Class I patients have no clinical signs of heart failure and have 30-day mortality rates of 2.8% and 6-month mortality rates of 5%; class II patients have basilar rales ± S3 gallop ± elevated jugular venous pressure with mortality rates close to 8.8% at 30 days and at 14.7% at 6 months. Class III is defined by frank pulmonary edema and class IV describes a patient in CS. In recent study of a group of patients with STEMI treated with primary PCI, advanced Killip class at presentation was associated with impaired epicardial and myocardial perfusion, a larger incidence of distal embolization, larger enzymatic infarct size, and lower predischarge ejection fraction.²⁰ Mortality rates were higher for patients with higher Killip classes at follow-up. Mortality data for class III and IV were combined to show an average 30-day mortality rate of 14.4% and average 6-month mortality rate of 23%.²¹

It is important to note, however, that there is a specific subset of patients with severe LV dysfunction who have nonhypotensive CS or evidence of peripheral hypoperfusion with preserved blood pressure.²² These patients should be identified, as they may require inotropic therapy and their mortality rates are similar to rates seen in patients with typical CS.

The presence of new cardiac murmurs may indicate mechanical complications of MI, such as a ventricular septal defect or acute mitral regurgitation, while distant heart sounds may indicate a pericardial effusion. In 30% of patients with CS, pulmonary edema will not be detected by auscultation.³

The EKG is a key tool in both the diagnosis and etiology of CS. Presence of ST-segment elevations consistent with a STEMI usually indicates the shock is cardiogenic and that the severity of the infarction is consistent with the level of clinical severity. Anterior ST-segment elevation suggests that LV acute pump failure is likely the cause of CS. If ST-segment elevations are present in the inferior leads, marked ST-segment elevations with reciprocal depressions should be present. A right-sided EKG may show elevation of RV leads indicative of RV infarction. A relatively normal EKG in a patient with clinical CS should indicate the possibility of myocarditis, especially if the patient has arrhythmias.

A high-quality chest radiography is usually performed to assess for signs of pulmonary edema, but may be most useful to guide initial care when signs suggest an alternative diagnosis, such as a widened mediastinum indicative of aortic dissection. Other notable pathology may include cardiomegaly and pulmonary venous hypertension.

An echocardiogram is indicated in a patient with suspected CS to aid both diagnosis and management and is a class I indication in the joint American College of Cardiology/American Heart Association (ACC/AHA) Practice Guidelines.²³ Echocardiography is noninvasive, low-risk, and can help to provide data regarding cardiac chamber size, both LV and RV function, valvular structure and motion, atrial size, and the anatomy of the pericardial space. Transthoracic echocardiography can be used to ascertain overall and regional systolic function as well as diastolic function. Echocardiography also helps in determining the presence or absence of mechanical causes of shock, such as papillary muscle rupture, acute ventricular septal defect, and free wall rupture. It can also be used to interrogate for the degree of mitral regurgitation, presence of RV infarction, and to exclude other causes of shock, such as cardiac tamponade, pulmonary embolism, and valvular stenosis.

If the history, physical examination, chest radiograph, and echocardiography clearly demonstrate systemic hypoperfusion, low cardiac output and elevation of left atrial pressure, pulmonary artery pressure, and right atrial pressure, a right heart catheterization (RHC) may not be necessary for diagnosis, but therapy with vasopressors and inotropes is best optimized using hemodynamic measurements unless titration is planned. A RHC may be useful both for the diagnosis and optimization of therapy. Diagnosis of CS using RHC includes a cardiac index < 2.2 L/min/m² in the setting of a PCWP > 18 mm Hg. Right heart catheterization helps to exclude other causes of shock, such as volume depletion and septic shock, and helps to diagnose RV infarction and mechanical complications. A step-up in oxygen saturation between the right atrium and right ventricle can diagnose a ventricular septal defect, and the presence of large V waves in the PCWP waveform can indicate the presence of acute severe mitral regurgitation. Right ventricular infarction can be inferred when PCWP is normal but right-sided filling pressures are notably elevated.

To date, no randomized trials have evaluated the use of indwelling RHCs specifically in patients with CS, but 96% of patients in the landmark SHOCK trial had RHCs (because their use was mandated).²⁴ Right heart catheterization is useful for obtaining indices of cardiac output to guide the use of inotropic agents, and obtaining filling pressures to guide the use of both vasopressors and vasodilators. Other methods of obtaining these indices are acceptable, but measurement of hemodynamic parameters allows titration of inotropic and vasopressor agents to the minimum dosage required to achieve therapeutic goals, while minimizing increases in oxygen demands and arrhythmias.

Initial stabilization of patients with suspected CS should consist of venous access, supplemental oxygen and mechanical ventilation, if necessary, and continuous telemetry monitoring. If the etiology is likely to be acute myocardial ischemia, aspirin and heparin should be given. β-Blockade and ACE inhibitors should be avoided in patients with tenuous hemodynamic status.³

An initial assessment of fluid status and systemic perfusion should be made. Patients are commonly diaphoretic and relative hypovolemia may be present. In the original description of hemodynamic subsets in MI, approximately 20% of patients had both a low cardiac index and low PCWP; most had reduced stroke volume and compensatory tachycardia.²⁵ Some of these patients may benefit from judicious fluid replacement with predetermined boluses titrated to clinical endpoints. Ischemia produces diastolic as well as systolic dysfunction, and thus elevated filling pressures may be necessary to maintain stroke volume in some patients with CS.

If hypotension is unresponsive to fluid challenges, vasopressors, such as dopamine and norepinephrine, may be initiated and should be titrated to clinical indices of perfusion and mixed venous oxygen saturation. Dopamine and norepinephrine are considered first-line treatment for hypotension in this situation. Invasive hemodynamic monitoring with an arterial line and right heart catheter are advisable during titration of inotropes.

Dopamine acts as both an inotrope (particularly between 3–10 μg/kg/min) and a vasopressor. Norepinephrine acts primarily as a vasoconstrictor, has a mild inotropic effect, and increases coronary flow. Vasopressin, which causes vasoconstriction, has a neutral or slightly depressant effect on cardiac output, and increases vascular sensitivity to norepinephrine; it may be added to catecholamines if needed. If inadequate tissue perfusion remains, inotropic therapy should be initiated and consideration should be given to placement of an intraaortic balloon pump (IABP).

Dobutamine, a selective β1-adrenergic receptor agonist, can improve myocardial contractility and increase cardiac output, and is the first-choice agent for patients with a low-output syndrome without shock.²⁶ Dobutamine may exacerbate hypotension in some patients due to its vasodilatory effects and can precipitate tachyarrhythmias. Milrinone, a phosphodiesterase inhibitor, increases cardiac contractility by increasing intracellular cyclic adenosine monophosphate. Milrinone has fewer chronotropic and arrhythmogenic effects than catecholamines, but analysis of data from the Outcomes of a Prospective Trial of Intravenous Milrinone for Exacerbations of Chronic Heart Failure (OPTIME-CHF) trial suggest worsened outcomes in patients with ischemic cardiomyopathy, although patients with shock were not included in this study.²⁷ Milrinone has the potential to cause hypotension and has a long half-life; in patients with tenuous clinical status, its use is often reserved for situations in which other agents have proven ineffective.³ In some cases, milrinone can be combined with dobutamine for an increased inotropic effect. Standard administration of milrinone calls for a loading dose followed by an infusion, but most clinicians eschew the loading dose in patients with marginal blood pressure.

Levosimendan, a calcium sensitizer, has both inotropic and vasodilatory properties, and does not increase myocardial oxygen consumption. Several relatively small studies have shown hemodynamic benefits with levosimendan in CS after MI,²⁸ one suggesting a better hemodynamic effect than dobutamine,²⁹ but survival benefits with use of levosimendan have not been shown in either CS or acute heart failure.³⁰ Levosimendan has the potential to cause hypotension and, thus, should be used with some caution in patients with CS. However, the current data suggest that it is no worse than dobutamine and there is as much or more evidence supporting its safety and efficacy as there is for any other intravenous inotropic or vasodilator agent.

Intraaortic balloon pumps help to bridge patients through a critical time of shock but are not definitive therapy.³¹ They are inserted through the femoral artery into the aorta and are placed distal to the arch of the aorta but should be higher than the renal arteries so as not to obstruct flow to the kidneys. They should not be used in patients with moderate-to-severe aortic valve regurgitation or in patients with severe atherosclerotic aortoiliac disease. Intraaortic balloon pumps inflate during diastole when the coronary arteries fill and help to augment coronary flow. During systole, IABPs help to reduce afterload as the balloon deflates. These beneficial effects occur without causing an increased oxygen demand.

The use of IABPs for STEMI patients when CS is not quickly reversed with pharmacological therapy is a class I indication according to the ACC/AHA guidelines.²³ The guidelines also note that “the IABP is a stabilizing measure for angiography and prompt revascularization.”²³ Intraaortic balloon pump use has been associated with increased rates of survival in studies of acute MI.^32,33

In an acute STEMI setting, fibrinolytic therapy has been proven to restore patency to the culprit artery, which is known to decrease the likelihood of progression to CS. Once a patient has already developed CS, however, fibrinolytic therapy may not be as effective as if the patient had been treated with fibrinolytic prior to development of shock.³⁴ In the SHOCK registry, patients treated with fibrinolytic therapy had a lower in-hospital mortality rate than those who were not (54% vs 64%; P = 0.005), even after adjusting age and revascularization status (odds ratio, 0.70; P = 0.027).³³ It is, in fact, a class I recommendation, according to the ACC, that fibrinolytic therapy be administered to STEMI patients with CS who are unsuitable for further invasive care and do not have contraindications to fibrinolysis (level of evidence: B).²³

Data favors early revascularization for patients in the setting of CS. If a cardiac catheterization laboratory is available, then catheterization plus revascularization with angioplasty or coronary artery bypass grafting (CABG) is preferable.²³ If an institution has no catheterization laboratory, thrombolytic therapy plus IABP has good outcomes. The patient should then be quickly transferred to a hospital with catheterization and revascularization capabilities.

The issue of timing of revascularization in patients with CS secondary to MI was addressed in the SHOCK trial.²⁴ Patients with shock caused by LV failure complicating STEMI were randomly assigned to emergency revascularization (n = 152), accomplished by either CABG or angioplasty, or initial medical stabilization (n = 150), during which no coronary interventions or bypass surgery were allowed for 54 hours. Intraaortic balloon pumps were used in 86% of patients in both groups. The primary endpoint, all-cause mortality at 30 days, was 46.7% in the revascularization group and 56% in the medical therapy group, a difference that did not reach statistical significance (P = 0.11).²⁴ Planned follow-up, however, revealed a significant benefit from early revascularization at 6 months and 1 year (P < 0.03).³⁵ Subgroup analyses also revealed benefit in patients who were aged < 75 years, had a prior MI, and were randomly assigned < 6 hours from onset of infarction.²⁴ The authors concluded that long-term survival improved significantly in patients with CS who underwent early revascularization.

The SHOCK trial formed the basis of the current ACC/AHA guidelines regarding revascularization in CS, which states that “emergent coronary revascularization is the standard of care for CS due to pump failure (acute MI and shock).”²³ Early revascularization, either PCI or CABG, is recommended for patients aged < 75 years with ST-segment elevation or left bundle branch block who develop shock within 36 hours of MI and are suitable for revascularization that can be performed within 18 hours of shock unless further support is futile because of the patient’s wishes or contraindications/unsuitability for further invasive care (level of evidence: A).²³

Analysis of the SHOCK trial helps to define the indications for CABG in the setting of CS. Coronary artery bypass grafting should be the first-line therapy offered in cases of left main disease or triple vessel disease, and in cases in which the patient has sustained mechanical complications necessitating surgical repair. For patients with multivessel disease in the SHOCK trial, complete revascularization was achieved more frequently (87% vs 23%), although long-term mortality rates were similar in the CABG and PCI groups.³⁶

Mechanical support with left ventricular assist devices (LVADs) can interrupt the downward spiral of myocardial dysfunction, hypoperfusion, and ischemia in CS, allowing time for recovery of stunned or hibernating myocardium. In CS, after acute MI, percutaneous LVADs may be placed in the catheterization laboratory. Left ventricular assist devices can be used as bridges to cardiac transplantation or as bridges to recovery in patients with CS. Percutaneously implanted LVADs are used in cases of CS, during high-risk PCIs, in postcardiotomy shock, and in fulminant myocarditis. Two currently approved devices, the TandemHeart^® and the Impella^®, are placed through the femoral artery. Both devices offer complete cardiac support but do require adequate RV function; TandemHeart^® placement involves trans-septal puncture while the Impella^® does not. These devices augment cardiac output and blood pressure while decreasing myocardial oxygen demand. Known complications of percutaneous LVAD use include limb ischemia and bleeding.

Two recent trials compared the use of IABP with the TandemHeart^® for patients with CS^37,38 and another compared use of the Impella^® with IABP therapy;³⁹ the results of these trials were combined in a meta-analysis that included 100 patients.⁴⁰ Hemodynamic benefits for percutaneous LVADs compared with IABPs were found, as were higher cardiac indices and mean arterial pressures, and lower PCWPs. However, LVAD use showed no mortality benefit over IABPs at 30 days.⁴⁰

For patients with end-stage heart failure and refractory shock, a variety of surgically placed assist devices can be employed for circulatory support. These devices retrieve blood from the LV apex and use a pumping device, either continuous or pulsatile, to return the blood to the ascending aorta. Full consideration of these devices is beyond the scope of this article; in CS, they are usually used as a bridge to recovery or transplantation, although in other contexts they may be used as destination therapy.

Cardiogenic shock remains a prevalent and dangerous syndrome that requires accurate and efficient diagnosis. Treatment has advanced so that the condition, once regarded as uniformly fatal, is now proving treatable. The potential for reversal of myocardial dysfunction with revascularization provides the rationale for supportive therapy to maintain coronary and tissue perfusion until more definitive revascularization measures can be undertaken. Application of a thorough understanding of the essentials of pathophysiology, diagnosis, and treatment of CS can allow for expeditious management and improved outcomes.

Hello!


Back to the table of contents for the February 2010 issue