Stat | Lytes | Drugs | ID | Heart | Lungs | Kidneys | GI | Rheum | Heme-Onc | Endo | Neuro | Derm | Misc. | Resource | Home
REVIEW - Cardiogenic Shock
Steven M. Hollenberg, MD; Clifford J. Kavinsky, MD, PhD; and Joseph E. Parrillo, MD
Cardiogenic shock is a state of inadequate tissue perfusion due to cardiac
dysfunction, most commonly caused by acute myocardial infarction.
The key to achieving a good outcome is an organized approach that includes rapid diagnosis and prompt initiation of therapy to maintain blood pressure and cardiac output.
Expeditious coronary revascularization is crucial.
Conclusions: Improved understanding of the pathophysiology of shock and myocardial infarction has led to improved treatment. If cardiogenic shock is managed with rapid evaluation and prompt initiation of supportive measures and definitive therapy, outcomes can be improved.
Ann Intern Med. 1999; 131: 47-59.
The clinical definition of cardiogenic shock is decreased cardiac output and evidence of tissue hypoxia in the presence of adequate intravascular volume. Hemodynamic criteria are sustained hypotension (systolic blood pressure < 90 mm Hg for at least 30 minutes) and a reduced cardiac index (< 2.2 L/min per m2) in the presence of elevated pulmonary capillary occlusion pressure ( > 15 mm Hg) (2).
Circulatory shock is diagnosed at the bedside by observing hypotension and clinical signs indicating poor tissue perfusion, including oliguria; clouded sensorium; and cool, mottled extremities. Cardiogenic shock is diagnosed after documentation of myocardial dysfunction and exclusion or correction of such factors as hypovolemia, hypoxia, and acidosis.
The most common cause of cardiogenic shock is extensive acute myocardial infarction, although a smaller infarction in a patient with previously compromised left ventricular function may also precipitate shock.
Cardiogenic shock can also be caused by mechanical complicationssuch as acute mitral regurgitation, rupture of the interventricular septum, or rupture of the free wallor by large right ventricular infarctions. Other causes of cardiogenic shock include myocarditis, end-stage cardiomyopathy, myocardial contusion, septic shock with severe myocardial depression, myocardial dysfunction after prolonged cardiopulmonary bypass, valvular heart disease, and hypertrophic obstructive cardiomyopathy (Table 1).
In a recent report of the SHOCK (SHould we emergently revascularize Occluded Coronaries for shocK) trial registry of 1160 patients with cardiogenic shock (1, 9),
Patients may have cardiogenic shock at initial presentation, but shock often evolves over several hours (10, 11). In the SHOCK trial registry (1, 9), 75% of patients developed cardiogenic shock within 24 hours after presentation; the median delay was 7 hours from onset of infarction. Results from the GUSTO trial (8) are similar: Among patients with shock, 11% were in shock on arrival and 89% developed shock after admission.
Among patients with myocardial infarction, shock is more likely to develop in those who are elderly, are diabetic, and have anterior infarction (10, 12-14). Patients with cardiogenic shock are also more likely to have histories of previous infarction, peripheral vascular disease, and cerebrovascular disease (13, 14). Decreased ejection fractions and larger infarctions (as evidenced by higher cardiac enzyme levels) are also predictors of the development of cardiogenic shock (13, 14).
Cardiogenic shock is most often associated with anterior myocardial infarction. In the SHOCK trial registry (1, 9),
These findings were consistent with those in other series (15). Angiographic evidence most often demonstrates multivessel coronary disease (left main occlusion in 29% of patients, three-vessel disease in 58% of patients, two-vessel disease in 20% of patients, and one-vessel disease in 22% of patients) (9).
Reversible Myocardial Dysfunction
This reversible myocardial dysfunction can be described in two main categories: stunning and hibernation.
Clinical Assessment and Initial Management
Cardiogenic shock is an emergency. The clinician must initiate therapy before shock irreversibly damages vital organs; at the same time, he or she must perform the clinical assessment required to understand the cause of shock and target therapy to that cause. A practical approach is to make a rapid initial evaluation on the basis of a limited history, physical examination, and specific diagnostic procedures (Figure 3).
Cardiogenic shock is diagnosed after documentation of myocardial dysfunction and exclusion of alternative causes of hypotension, such as hypovolemia, hemorrhage, sepsis, pulmonary embolism, tamponade, aortic dissection, and preexisting valvular disease.
Patients with shock are usually ashen or cyanotic and can have cool skin and mottled extremities. Cerebral hypoperfusion may cloud the sensorium. Pulses are rapid and faint and may be irregular if arrhythmia is present. Jugular venous distention and pulmonary rales are usually present, although their absence does not exclude the diagnosis. A precordial heave resulting from left ventricular dyskinesis may be palpable. The heart sounds may be distant, and third or fourth heart sounds, or both, are usually present. A systolic murmur of mitral regurgitation or ventricular septal defect may be heard, but these complications may also occur without an audible murmur.
The initial approach to the patient in cardiogenic shock should include
Although it has been convincingly demonstrated that thrombolytic therapy reduces mortality rates in patients with acute myocardial infarction (4, 63-65), the benefits of this therapy in patients with cardiogenic shock are less certain. It is clear that thrombolytic therapy can reduce the likelihood of subsequent development of shock after initial presentation (7, 8, 63, 64). This is important because most patients develop cardiogenic shock more than 6 hours after hospital presentation (1, 8, 9).
Nonetheless, no trials have demonstrated that thrombolytic therapy reduces mortality rates in patients with established cardiogenic shock.
The failure of thrombolytic therapy to improve survival in patients with cardiogenic shock may seem paradoxical in light of evidence that the absolute reduction in mortality rates resulting from use of thrombolytic agents is greatest in persons who are at highest risk at presentation. The meta-analysis performed by the Fibrinolytic Therapy Trialists Collaborative Group (68) demonstrated a reduction in mortality rate from 36.1% to 29.7% when thrombolytic therapy was used in patients with initial systolic blood pressures less than 100 mm Hg. In patients with initial heart rates greater than 100 beats/min, the mortality rate decreased from 23.8% to 18.9%. However, most patients in these subgroups did not meet the criteria for cardiogenic shock.
Intra-Aortic Balloon Pumping
Intra-aortic balloon pumping (IABP) reduces systolic afterload and augments diastolic perfusion pressure, increasing cardiac output and improving coronary blood flow (75). Intra-aortic balloon pumping is efficacious for initial stabilization of patients with cardiogenic shock (76). Small randomized trials in the prethrombolytic era, however, did not show that IABP alone increases survival (77, 78). Intra-aortic balloon pumping alone does not substantially improve blood flow distal to a critical coronary stenosis (79).
Intra-aortic balloon pumping is probably not best used as an independent method to treat cardiogenic shock. It may, however, be an essential support mechanism to allow definitive therapeutic measures to be undertaken. In the GUSTO trial, patients who presented with shock and had early IABP placement showed a trend toward lower mortality rates, even after exclusion of patients who had revascularization (7, 80). A similar trend was seen in the SHOCK trial registry (1), although it did not persist after adjustment for age and catheterization. Several observational studies have also suggested that IABP can improve outcome in patients with shock, although revascularization procedures are a confounding factor in these studies (81-84). Intra-aortic balloon pumping has been shown to decrease reocclusion and cardiac events after emergency angioplasty for acute myocardial infarction (85, 86).
In hospitals without direct angioplasty capability, stabilization with IABP and thrombolysis followed by transfer to a tertiary care facility may be the best management option. Intra-aortic balloon pumping may be a useful adjunct to thrombolysis in this setting by increasing drug delivery to the thrombus, improving coronary flow to other regions, preventing hypotensive events, or supporting ventricular function until areas of stunned myocardium can recover (87). Two retrospective studies (83, 84) have shown that patients with cardiogenic shock who were treated in a community hospital with IABP placement followed by thrombolysis had improved in-hospital survival and improved outcomes after subsequent transfer for revascularization, although selection bias is clearly a confounding factor. The role of IABP as an adjunct to thrombolytic agents in the community hospital is being addressed in the ongoing randomized, multicenter Thrombolysis And Counterpulsation To Improve Cardiogenic Shock Survival (TACTICS) trial.
Pathophysiologic considerations and extensive retrospective data favor aggressive mechanical revascularization for patients with cardiogenic shock caused by myocardial infarction. Recently, a landmark study (the SHOCK trial ) presented data from a randomized, controlled trial.
Direct Coronary Angioplasty
Reestablishment of brisk (TIMI [Thrombolysis in Myocardial Infarction] grade 3) flow in the infarct-related artery is an important determinant of left ventricular function and survival after myocardial infarction (69). Direct percutaneous transluminal coronary angioplasty (PTCA) can achieve TIMI grade 3 flow in 80% to 90% of patients with myocardial infarction (89-91), compared with rates of 50% to 60% 90 minutes after thrombolytic therapy (69, 92). In the Primary Angioplasty in Myocardial Infarction (PAMI) trial (89), which compared direct angioplasty with thrombolytic therapy, a mortality benefit for PTCA (in-hospital mortality rate, 2.0% compared with 10.4%; P = 0.01) was seen in high-risk patients (age > 70 years, large anterior myocardial infarction, heart rate >100 beats/min). Therefore, patients with cardiogenic shock are candidates for direct angioplasty. In addition to improving wall motion in the infarct territory, increased perfusion of the infarct zone has been associated with augmented contraction of remote myocardium, possibly caused by recruitment of collateral blood flow (16).
Several retrospective trials (15, 71, 93-114) have examined the effect of angioplasty on mortality rates in patients with cardiogenic shock (Table 2). Researchers have consistently found that patients with successful reperfusion have much better outcomes than those without successful reperfusion. In a subgroup analysis of the 2972 patients with cardiogenic shock, the GUSTO-1 trial (8) showed that the 30-day mortality rate was significantly lower in patients who had angioplasty (43% compared with 61% for patients with shock on arrival; 32% compared with 61% for patients who developed shock). In this trial, patients treated with an aggressive strategy (coronary angiography performed within 24 hours of shock onset with revascularization by PTCA or bypass surgery) had a significantly lower mortality rate (38% compared with 62%) (71). This benefit was present even after adjustment for baseline characteristics (71) and persisted for up to 1 year (115).
The role of newer developments in PTCA for acute myocardial infarction in patients with cardiogenic shock remains to be defined. Recent reports have suggested that placement of coronary stents may improve outcome, after either failed or suboptimal PTCA (116) or as a primary approach (117, 118). The PAMI stent pilot trial (119) and the Intracoronary Stenting and Antithrombotic Regimens (ISAR) trial (120) have recently shown that primary stenting is feasible in patients with acute myocardial infarction; TIMI grade 3 flow is restored in more than 90% of patients, and short-term outcome is good. Data in patients with cardiogenic shock are more sparse. A recent study of direct PTCA in patients with shock (114) reported a success rate of 94%, with placement of stents in 47% of patients; the in-hospital mortality rate was 26%. Another study of stenting for failed angioplasty in patients with cardiogenic shock reported a mortality rate of 27% (121).
The role of adjunctive antiplatelet therapy is also evolving. Platelet glycoprotein IIb/IIIa antagonists have been shown to improve short-term clinical outcomes after angioplasty, especially in patients at high risk for complications (122, 123). Published experience with IIb/IIIa receptor inhibition in patients with cardiogenic shock is thus far limited to case reports (124), but extrapolation from other settings suggests that they may play an important adjunctive role in patients with shock who undergo angioplasty (125).
Coronary Artery Bypass Surgery
Many trials have reported favorable outcomes for patients with cardiogenic shock who have coronary artery bypass surgery (81, 121, 126, 127). Left main and three-vessel coronary disease are common in patients with cardiogenic shock (1, 9, 69), and the potential contribution of ischemia in the noninfarct zone to myocardial dysfunction in patients with shock would support complete revascularization. Nonetheless, the logistic and time considerations involved in mobilizing an operating team, the high surgical morbidity and mortality rates, and the generally favorable results of percutaneous interventions discourage routine bypass surgery for these patients. In DeWood and colleagues' series (81), IABP support was used successfully as a bridge to coronary artery bypass surgery. The roles of other supportive measures, such as emergency cardiopulmonary bypass surgery (128), remain to be defined.
These studies of revascularization in patients with cardiogenic shock were retrospective and uncontrolled. Selection bias is clearly present because patients selected for revascularization tend to be younger, less critically ill, and more likely to receive IABP support; they also tend to have less comorbidity (1, 8, 71). In addition, patients whose conditions deteriorate before planned revascularization is performed are counted in the nonrevascularized group. In the SHOCK trial registry (1), not only was the mortality rate in patients selected for cardiac catheterization lower than in those not selected (51% compared with 85%), but the mortality rate was also lower in catheterized patients who did not undergo revascularization (58%). A small randomized study, the Swiss Multicenter Evaluation of Early Angioplasty for SHock (SMASH) trial (129), showed no significant difference in mortality rate between patients who were randomly assigned to undergo angioplasty and those who were randomly assigned to receive medical treatment (69% compared with 78%), although the trial was stopped early because of difficulties in patient recruitment (129).
The recently presented multicenter SHOCK trial (88) is a landmark study because it contains the only data from a randomized, controlled study addressing revascularization in patients with cardiogenic shock. The SHOCK trial randomly assigned patients with cardiogenic shock to receive optimal medical managementincluding IABP and thrombolytic therapyor cardiac catheterization with revascularization using PTCA or coronary artery bypass grafting. The results of the trial were recently presented at the annual meeting of the American College of Cardiology (88).
The trial was powered to detect a 20% absolute decrease in 30-day all-cause mortality rates; 302 patients were enrolled. The mortality rate at 30 days was 46.7% in patients treated with early intervention and 56% in patients treated with initial medical stabilization; however, this difference did not reach statistical significance (absolute risk reduction, -9.3% [95% CI, -20.5 to 1.9%]; P = 0.11). At 6 months, the absolute risk reduction was 12% (54% compared with 66%). Preliminary analysis suggested that this difference was statistically significant. Subgroup analysis showed a substantial improvement in mortality rates in patients younger than 75 years of age at 30 days (41% compared with 57%) and 6 months (48% compared with 69%).
It is important to note that the controls (patients who received medical management) had a lower mortality rate than that reported in previous studies; this may reflect the aggressive use of thrombolytic therapy (64%) and IABP (86%) in these controls (88). These data provide indirect evidence that thrombolysis plus IABP may produce the best outcomes when cardiac catheterization is not immediately available.
In our judgment, the SHOCK trial was underpowered to detect the primary end point (30-day mortality). This may have been due to a lower mortality rate among controls than might have been expected (88). The improved survival with revascularization at 6 months and in patients younger than 75 years of age strongly supports the superiority of a strategy of early revascularization in most patients with cardiogenic shock (Figure 3).
Right Ventricular Infarction
Right ventricular infarction occurs in up to 30% of patients with inferior infarction and is clinically significant in 10% (130). Patients present with hypotension, elevated neck veins, and clear lung fields. Diagnosis is made by identifying ST-segment elevation in right precordial leads or characteristic hemodynamic findings on right-heart catheterization (elevated right atrial and right ventricular end-diastolic pressures with normal to low pulmonary artery occlusion pressure and low cardiac output). Echocardiography can show depressed right ventricular contractility (41). Patients with cardiogenic shock on the basis of right ventricular infarction have a better prognosis than those with left-sided pump failure (130). This may be due in part to the fact that right ventricular function tends to return to normal over time with supportive therapy (131), although such therapy may need to be prolonged.
Supportive therapy for patients with right ventricular infarction begins with maintainance of right ventricular preload with fluid administration. In some cases, however, fluid resuscitation may increase pulmonary capillary occlusion pressure but may not increase cardiac output, and overdilation of the right ventricle can compromise left ventricular filling and cardiac output (131). Inotropic therapy with dobutamine may be more effective in increasing cardiac output in some patients, and monitoring with serial echocardiography may also be useful to detect right ventricular overdistention (131). Maintenance of atrioventricular synchrony is also important in these patients to optimize right ventricular filling (41). For patients with continued hemodynamic instability, intra-aortic balloon pumping may be useful, particularly because elevated right ventricular pressures and volumes increase wall stress and oxygen consumption and decrease right coronary perfusion pressure, exacerbating right ventricular ischemia.
Reperfusion of the occluded coronary artery is also crucial. A recent study using direct angioplasty (132) showed that restoration of normal flow resulted in dramatic recovery of right ventricular function and a mortality rate of only 2%, whereas unsuccessful reperfusion was associated with persistent hemodynamic compromise and a mortality rate of 58%.
Acute Mitral Regurgitation
Ischemic mitral regurgitation is usually associated with inferior myocardial infarction and ischemia or infarction of the posterior papillary muscle, which has a single blood supply (usually from the posterior descending branch of a dominant right coronary artery) (133). Papillary muscle rupture usually occurs 2 to 7 days after acute myocardial infarction; it presents dramatically with pulmonary edema, hypotension, and cardiogenic shock. When a papillary muscle ruptures, the murmur of acute mitral regurgitation may be limited to early systole because of rapid equalization of pressures in the left atrium and left ventricle. More important, the murmur may be soft or inaudible, especially when cardiac output is low (134).
Echocardiography is extremely useful in the differential diagnosis, which includes free-wall rupture, ventricular septal rupture, and infarction extension pump failure. Hemodynamic monitoring with pulmonary artery catheterization may also be helpful. Management includes afterload reduction with nitroprusside and intra-aortic balloon pumping as temporizing measures. Inotropic or vasopressor therapy may also be needed to support cardiac output and blood pressure. Definitive therapy, however, is surgical valve repair or replacement, which should be undertaken as soon as possible because clinical deterioration can be sudden (134, 135).
Ventricular Septal Rupture
Patients who have ventricular septal rupture have severe heart failure or cardiogenic shock, with a pansystolic murmur and a parasternal thrill. The classic finding is a left-to-right intracardiac shunt (a "step-up" in oxygen saturation from right atrium to right ventricle). On pulmonary artery occlusion pressure tracing, ventricular septal rupture can be difficult to distinguish from mitral regurgitation because both can produce dramatic "V" waves. The diagnosis is most easily made with echocardiography.
Rapid stabilizationusing intra-aortic balloon pumping and pharmacologic measuresfollowed by surgical repair is the only viable option for long-term survival. The timing of surgery is controversial, but most experts now suggest that operative repair should be done early, within 48 hours of the rupture (135-137).
Ventricular free-wall rupture usually occurs during the first week after myocardial infarction; the classic patient is elderly, female, and hypertensive. The early use of thrombolytic therapy reduces the incidence of cardiac rupture, but late use may increase the risk. Free-wall rupture presents as a catastrophic event with a pulseless rhythm. Salvage is possible with prompt recognition, pericardiocentesis to relieve acute tamponade, and thoracotomy with repair (138).
Reversible Myocardial Dysfunction
In addition to hibernating and stunned myocardium, potentially reversible causes of myocardial dysfunction include sepsis-associated myocardial depression, myocardial dysfunction after cardiopulmonary bypass, and inflammatory myocarditis (139). In sepsis and, to some extent, in myocarditis, myocardial dysfunction seems to result from the effects of inflammatory cytokines, such as tumor necrosis factor and interleukin-1 (140-142). Myocardial dysfunction can be self-limited or fulminant, with severe congestive heart failure and cardiogenic shock (139). In the latter situation, cardiovascular support with a combination of inotropic agents (such as dopamine, dobutamine, or milrinone) and IABP may be required for hours or days to allow sufficient time for recovery. If these measures fail, mechanical circulatory support with left ventricular assist devices can be considered (143). These devices can be used as a bridge to cardiac transplantation in eligible patients or as a bridge to myocardial recovery; functional improvement with such support can be dramatic (143).
Mortality rates in patients with cardiogenic shock remain frustratingly high (50% to 80%). The pathophysiology of shock involves a downward spiral: Ischemia causes myocardial dysfunction, which in turn worsens ischemia. Areas of nonfunctional but viable myocardium can also cause or contribute to the development of cardiogenic shock. The key to a good outcome is an organized approach with rapid diagnosis and prompt initiation of therapy to maintain blood pressure and cardiac output. Expeditious coronary revascularization is crucial. When available, emergency cardiac catheterization and revascularization with angioplasty or coronary surgery seem to improve survival and represent standard therapy at this time. In hospitals without direct angioplasty capability, stabilization with IABP and thrombolysis followed by transfer to a tertiary care facility may be the best option. The SHOCK multicenter randomized trial (88) provides important data that help clarify the appropriate role and timing of revascularization in patients with cardiogenic shock.
Requests for Reprints: Steven M. Hollenberg, MD, Section of Cardiology, Rush-Presbyterian-St. Luke's Medical Center, 1725 West Harrison Street, Suite 1159, Chicago, IL 60612.