The incidence of cardiogenic shock has not changed in the reperfusion era. The National Registry of Myocardial Infarction (NRMI) from January 1995 to May 2004 at 775 US hospitals with revascularization capability included 293,633 patients with ST-elevation myocardial infarction (MI), 25,311 (8.6%) of whom had cardiogenic shock. Presentation with cardiogenic shock occurred in 7,356 patients (2.9%).1 The mortality rate in this study fell from 60.3 to 47.9% over 10 years. In-hospital mortality from cardiogenic shock from hospitals in a single community—Worcester, MA—also fell by 10% for the period 1975–1990 from approximately 80% to approximately 60% during 1995–1997.2 In the GUSTO I trial, 1,891 patients with cardiogenic shock were enrolled in the US; 30-day mortality was 49.6% and mortality at 11 years was 72.2%.3 Annual mortality rates of 2.4% per year for 30-day survivors following shock were similar to those without shock.
Much of our understanding of the benefit of emergency revascularization in managing cardiogenic shock comes from the SHOCK (SHould we emergently revascularize Occluded Coronaries in cardiogenic shocK?) trial,4 a landmark study that also raised intriguing questions. The SHOCK trial was a randomized, controlled study designed to examine whether emergency revascularization (ERV)—either angioplasty or bypass surgery—was superior to a strategy of initial stabilization by means of intensive medical therapy (IMS). IMS included a recommendation for intra-aortic balloon counterpulsation, fibrinolytic therapy if feasible, and delayed revascularization if clinically warranted. The initial assumption in the trial, designed in 1992, was that emergency revascularization would result in a 20% absolute reduction in the primary end-point of overall mortality at 30 days.4,5 At the end of 5.5 years, the trial had randomized 152 patients to ERV, with percutaneous coronary intervention (PCI) for one-vessel disease and emergency coronary artery bypass grafting (CABG) or staged percutaneous coronary intervention (PCI) for multivessel disease, and 150 to the IMS strategy. The primary end-point of overall mortality at 30 days was negative—46.7% ERV versus 56.0% IMS (p=0.11)—but at six months (a secondary end-point of the trial) mortality was 50.3% for ERV versus 63.1% for IMS (p=0.027).4 The survival advantage of an absolute 13 percentage points persisted at one year (46.7 versus 33.6%; p<0.03) and through six years (32.8 versus 19.6%; p<0.03).6,7Figure 1 shows one-year survival by treatment assignment in the SHOCK trial when left ventricular ejection fraction was used as an independent baseline variable. Of note, most survivors improved after discharge and were in New York Heart Association (NYHA) class I or II at six and 12 months after discharge (see Figure 2).8
Left Ventricular Function in SHOCK
An interesting observation from the echocardiographic substudy of the SHOCK trial was that the average ejection fraction was depressed (30%) but there was a wide range of ejection fractions and left ventricular sizes noted,9 nonetheless not reaching the very low levels of ejection fraction and high left ventricular volumes commonly seen in many chronic heart failure patients. On multivariable analysis in this echo SHOCK substudy,9 ejection fraction was the only independent predictor of survival apart from mitral regurgitation. A survival benefit for the ERV strategy was observed at all levels of ejection fraction and mitral regurgitation.
Angiographic Predictors of Survival
In an angiographic analysis10 (including 147 of 152 ERV patients and 100 of 150 IMS patients), the predictors of one-year mortality included ejection fraction, culprit vessel (worse outcome with left anterior descending artery [LAD] than right coronary artery [RCA]), number of diseased vessels (only in the IMS group), and initial TIMI flow grade of the culprit artery (only in the ERV group). In the 82 patients randomized to ERV who had PCI performed,10,11 restoration of TIMI-3 flow was the major predictor of survival.
Systemic Vascular Resistance and SHOCK
Systemic vascular resistance was not elevated in the patients recruited to the SHOCK trial, with a very wide range of systemic vascular resistance measured (average of 1,350–1,400 dyne.s.cm-5 despite the use of vasopressors).12 Evidence of a systemic inflammatory response was frequently present: 54 of the 302 patients demonstrated signs of inflammation such as fever or leukocytosis.13 Of interest, these patients had a lower systemic vascular resistance measured at the time of randomization in the trial before clinical evidence of inflammation developed. This is consistent with the concept that the spectrum of inflammatory response, including complement activation, release of inflammatory cytokines, expression of inducible nitric oxide (NO) synthase (iNOS), and inappropriate vasodilation, may play an important pathophysiological role in cardiogenic shock.14–16 While the wide variability of systemic vascular resistance could be due to many factors, including the use of vasoconstrictors administered in 99% of patients or vasodilators, it suggests that the pathophysiology of cardiogenic shock may vary in different patients.
Conventionally, it is thought that maximal peripheral vasoconstriction inevitably accompanies low cardiac output, so that cardiac output and arterial blood pressure will reflect the severity of cardiogenic shock. The recognition of low systemic vascular resistance in patients with cardiogenic shock has contributed to recognition of cardiac power output (the product of cardiac output and mean arterial pressure) as a good prognostic predictor.17 In 541 patients in the SHOCK trial registry, cardiac power output, cardiac power index, cardiac output, cardiac index, stroke volume, left ventricular work, left ventricular work index, stroke work, mean arterial pressure, systolic and diastolic blood pressure (all p<0.001), coronary perfusion pressure (p=0.002), ejection fraction (p=0.013), and pulmonary artery systolic pressure (p=0.047) were associated with in-hospital mortality. On multivariable analyses, cardiac power output was the strongest independent hemodynamic correlate of in-hospital mortality.17
From the therapeutic point of view, the concept that excessive NOS resulting in high levels of nitric oxide may cause inappropriate systemic vasodilatation, aggravating both systemic and coronary hypoperfusion and leading to decreased myocardial function, has led to the investigation of novel treatment approaches. The phase II dose-ranging SHOCK-2 trial (Should We Inhibit Nitric Oxide Synthase in Cardiogenic Shock 2)18 investigated the safety and tolerability of L-NG-monomethylarginine (L-NMMA) (tilarginine acetate injection) in patients with cardiogenic shock. Tilarginine was given as a bolus (0.15–1.5mg/kg) followed by five-hour infusion (0.15 to 1.5 mg/kg per hour), and resulted in modest improvements in hemodynamic parameters. The recently reported Tilarginine Acetate Injection in a Randomized International Study in 398 Unstable MI Patients With Cardiogenic Shock (TRIUMPH) trial19 was designed to test the effect of NOS inhibition with tilarginine on mortality due to persistent cardiogenic shock complicating MI despite an open infarct artery, either spontaneously (4%) or after a successful PCI (96%). There was no difference in any outcomes, including shock resolution, shock duration, heart failure status, and survival. After six months, mortality rates were 58% in the tilarginine group versus 59% in the placebo group.19
Early Mortality in Cardiogenic Shock
Despite the advantages of revascularization, the contemporary mortality rate remains approximately 50% in patients with cardiogenic shock. Half of the deaths occur within the first 48 hours,4,20 and the benefit of revascularization over medical therapy is not observed until after the first four or five days, as shown in the SHOCK trial (see Figure 3).
Treatment Algorithm for Cardiogenic Shock
A suggested treatment algorithm is shown in Figure 4.12 It is recommended to give abciximab, but there are no randomized trials in cardiogenic shock. Since the publication of the SHOCK trial in 1999,4 early revascularization has been a class I recommendation in guidelines from the American College of Cardiology and the American Heart Association for patients under 75 years of age.21
Role of Coronary Bypass Surgery in Patients with Shock
In the SHOCK trial, 36.7% (47 of 128 patients) in the ERV group had CABG as their modality of reperfusion, and one-year survival rates with CABG were comparable to those with PCI, despite the fact that patients undergoing CABG had more extensive coronary disease22 and PCI was performed immediately after diagnostic angiography, in contrast to an inevitable delay for CABG. For patients who develop cardiogenic shock several hours or days after AMI presentation, myonecrosis may or may not be ongoing and ischemic dysfunction from both hibernation and stunning may cause shock. Complete revascularization with CABG may offer the best relief from ischemia and the best approach to improve survival.
Timing of Intervention
As shown in Figure 5,23 mortality reduction as a benefit of reperfusion therapy is greatest in the first two to three hours after the onset of symptoms of acute MI, most likely a consequence of myocardial salvage. The exact duration of this critical early period may be modified by several factors, including the presence of functioning collateral coronary arteries, ischemic pre-conditioning, myocardial oxygen demands, and whether ischemia is sustained or intermittent. After this early period, the magnitude of the mortality benefit is much reduced, and as the mortality reduction curve flattens, time to reperfusion therapy is less critical.23
The high mortality in elderly patients with cardiogenic shock from left ventricular dysfunction complicating an acute ST-elevation MI is well recognized. In the NRMI analysis, mortality fell from 69.9 to 64.1% (p<0.001) in patients ≥75 years of age from 1995 to 2004, whereas in patients <75 years of age the fall was greater, from 55.8 to 39.5% (p<0.001).1 Also, in patients who underwent primary PCI from 1995 to 2004, survival was improved in patients <75 and not in those ≥75 years of age.1 This could be because higher-risk elderly patients had PCI over this period.
Over a median six-year follow-up of the SHOCK trial patients, the better long-term survival with ERV was consistent among multiple subgroups, including the elderly.7 At one year, there was an 18% absolute difference in survival in favor of ERV patients for those <75 years of age (51.6% for ERV versus 33.3% for IMS), whereas survival was lower in those >75 years of age who had ERV (20.8 versus 34.4%; p-value for age interaction 0.03). At six-year follow-up, this differential treatment effect for the age cut-off of 75 years was no longer statistically significant.7 The SHOCK investigators noted that the finding in the elderly was a subgroup finding that may have been due to chance and explained by an imbalance, with the elderly patients assigned to IMS having a higher baseline ejection fraction than those assigned to ERV.7 Opposite findings were found in the SHOCK Registry, which demonstrated a markedly lower adjusted risk for in-hospital mortality for ≥75 years of age who were clinically selected to undergo early revascularization;24 this is consistent with other registries.25,26 The 2004 revised American College of Cardiology/American Heart Association ST-elevation MI guidelines indicate that primary or rescue PCI or CABG surgery is reasonable for selected patients ≥75 years of age with cardiogenic shock (Class IIA recommendation).27 It is difficult to define in the acute setting which associated comortalities should exclude elderly patients from an early revascularization strategy, although in the SHOCK trial severe systemic illness, inability to gain vascular access, and unsuitability for revascularization were among the exclusion criteria. Expeditious diagnostic angiography and appropriate revascularization should be performed in patients ≥75 years of age without important comorbidities. It should be noted that there is little information to guide treatment in the extreme elderly (>85 years of age).
Patients with Right Ventricular Infarction
The SHOCK trial included only patients with cardiogenic shock from left ventricular failure. In the 933 patients collected in the non-randomized SHOCK trial registry,28 cardiogenic shock was clinically considered to be due to predominant right ventricular failure in 49 patients and left ventricular failure in 884 patients. Patients with predominant right ventricular shock were younger, with a lower prevalence of previous MI (25.5 versus 40.1%; p=0.047), anterior MI, and multivessel disease (34.8 versus 77.8%; p<0.001) and a shorter median time between the index MI and the diagnosis of shock (2.9 versus 6.2 hours; p=0.003) in comparison with patients with left ventricular shock. In patients with right ventricular shock, in-hospital mortality was 53.1% versus 60.8% for patients with left ventricular shock (p=0.296).
The beneficial effect of revascularization on mortality was not different between patients with right ventricular shock with revascularization (42.3 versus 65.2% medical treatment) compared with patients with left ventricular shock with revascularization (39.9 versus 78.3% medical treatment) (see Figure 6). Therefore, patients with shock due to RV infarction should be treated as aggressively as patients with shock due to left ventricular dysfunction.
Mortality from cardiogenic shock remains high with contemporary treatments. For patients initially treated with fibrinolysis, there should be a low threshold to perform rescue PCI if 50% ST segment resolution does not occur, indicating unsuccessful reperfusion,29 and with the earliest signs of shock such as tachycardia. New thrombus extraction catheters, modulation of metabolism, therapies to improve tissue perfusion, and developments in stem cell therapy may lead to improved patient outcomes. However, the best approach is prevention of shock, which requires early reperfusion therapy and recognition of lack of reperfusion in patients who receive fibrinolytic therapy.
Statement of Financial Disclosure: The authors have nothing to disclose financially or in the proprietary interest of the subject matter and all funding or financial support for the paper received through grants or sponsorship.
Both authors contributed to the analysis and interpretation of the data, and were involved in the drafting of the manuscript. Both authors have approved the final version for submission and have no conflicts of interest to disclose.