Electrocardiographic Evidence

Acute coronary syndromes are classified by the initial electrocardiogram (ECG), and patients are divided into three groups: those with ST-elevation myocardial infarction (STEMI), those without ST elevation but with enzyme evidence of myocardial damage (non–ST-elevation myocardial infarction, or NSTEMI), and those with unstable angina. Classification according to the presenting ECG coincides with current treatment strategies, since patients presenting with ST elevation benefit from immediate reperfusion. An ECG in patients with suspected acute coronary syndrome (ACS) should be obtained and interpreted within 10 minutes of presentation.²

Criteria for the diagnosis of STEMI include^1–3:

A number of potential pitfalls can contribute to misinterpretation of the ECG. Many conditions can mimic STEMI and lead to false positives. An early repolarization pattern with ≤ 3 mm ST elevation in leads V₁ to V₃ can be seen in healthy individuals, usually young men. Preexcitation, bundle branch block, pericarditis, pulmonary embolism, subarachnoid hemorrhage, metabolic disturbances such as hyperkalemia, hypothermia, and left ventricular (LV) aneurysm can be associated with ST elevation in the absence of acute myocardial ischemia. On the other hand, some conditions can lead to false negatives, including prior myocardial infarction (MI), paced rhythm, and LBBB when acute ischemia is not recognized. These pitfalls are common in the real world and in large clinical trials. For example, when ECGs from the GUSTO IIB trial were reviewed by expert readers at a core laboratory, 15% of patients with STEMI were found to have been misclassified as NSTEMI, and these patients had a 21% higher mortality rate.⁴

“Nondiagnostic” ECGs are common in the setting of acute MI. Up to 18% of patients subsequently determined to have MI have a normal ECG, and an additional 25% have nonspecific changes. These nondiagnostic ECG findings may be due to occlusion of small vessels only or to insensitivity of the 12-lead ECG to ischemia in the lateral or posterior LV territory. Visualization in the horizontal plane can be extended laterally and posteriorly by the addition of leads V₇ to V₉ and rightward by the addition of V_4R and V_5R. Systematic 15-lead ECG to include V_4R, V₈, and V₉ has been suggested to increase the sensitivity of diagnosing ST elevation from 47% to 59% without decreasing specificity.⁵ In fact, an 80-lead body surface mapping system has been shown to increase sensitivity and specificity of ECG diagnosis of ischemia, but challenges with rapid application at the bedside remain.⁶ If ischemia is strongly suspected, but changes are not seen on standard leads, obtaining an ECG with additional leads should be considered.³

ST-segment depression on ECG identifies patients with ACS at high risk. In the TIMI risk score, which has been shown to predict the likelihood of death and ischemic events, ST-segment changes, along with advanced age and prior coronary artery disease, show the strongest association with severe epicardial disease.⁷

The significance of T-wave changes is directly related to the pretest probability of disease. Large studies in asymptomatic patients show that most T-wave changes are nonspecific. In the coronary care unit, however, 87% of patients with only T-wave inversions across the precordium will have a significant left anterior descending (LAD) coronary artery stenosis by angiography. Among patients presenting to the emergency department with ACS, those with isolated T-wave changes have lower risk than those with ST depression but higher risk than those with a normal ECG.⁸

Cardiac Biomarkers

With cardiac cell death, proteins are released into the blood, and detection of these proteins has played a key role in establishing the diagnosis of ACS, risk stratification, and prediction of outcome. Beginning early in the 1970s, creatine kinase (CK) and its isoenzyme, MB, became the biomarkers of choice to establish myocardial injury and infarction. The sensitivity of CK-MB for diagnosis of MI at 6 hours is 91%, but at 2 and 4 hours, sensitivity is only 21% and 46%.⁹ The poor performance of CK-MB early in the course of MI led to the continued search for biomarkers that could diagnose MI early. Myoglobin was a contender for just such a role, because serum levels increase earlier than CK-MB, but the degree to which early sensitivity is increased is uncertain, and myoglobin lacks specificity.¹⁰

These biomarkers have now been superseded by troponin T and I, parts of the troponin-tropomyosin complex in cardiac myocytes. Troponin elevations are highly specific for myocardial cellular injury, except for infrequent false positives due to fibrin interference or cross-reacting antibodies.¹¹ Troponin is also much more sensitive than CK-MB because of its higher concentration in cardiac muscle; minor cardiac injury can elevate levels.¹¹ Even small increases in circulating troponin values correlate with adverse outcomes in the short and long term.¹¹ In non–ST-elevation ACS, elevated troponin levels not only predict increased risk but also identify the patients most likely to benefit from more aggressive antiplatelet strategies using IIb/IIIa inhibitors, use of low-molecular-weight heparin, and an early invasive strategy with coronary angiography and revascularization when appropriate.³

The challenge for the clinician, and in particular the intensivist, is that while elevation of serum troponin concentration is highly specific for myocardial cell damage, not all of that damage is a consequence of rupture of an atherosclerotic plaque. Other causes of elevated troponin, many of which are common in the intensive care unit (ICU), are listed in Table 28-2.

TABLE 28-2 Nonischemic Conditions Commonly Associated with Elevated Cardiac Troponin

Troponin release in critically ill patients may not always represent myocardial cell death. Endotoxin, cytokines, and other inflammatory mediators, along with catecholamines and conditions such as hypotension, therapy with inotrope agents, or hypoxia, can cause the breakdown of cytoplasmic troponin into smaller fragments that can pass through endothelial monolayers and subsequently be detected by sensitive assays for troponin.¹² Thus, detectable circulating troponin levels, although they usually emanate from myocardial cells, may not always represent either irreversible cell death or myocardial ischemia. Renal dysfunction is another factor associated with elevated circulating troponin levels, and both the sensitivity and specificity of this biomarker is decreased in this population.

Regardless of cause, it is clear that elevation of serum troponin levels is associated with worsened outcomes, both in and out of the ICU, even after adjustment for severity of disease.¹³ What is less clear is whether myocardial dysfunction represents the proximate cause of the worsened prognosis. It is often difficult to exclude ischemia in critically ill patients, but in a study of patients with septic shock, troponin predicted mortality, even among patients without flow-limiting coronary lesions (as assessed by stress echocardiography or autopsy).¹⁴

A further difficulty in the ICU is that patients may not experience classic symptoms of ischemia or may be unable to report them. Despite this potentially confounding factor, it is useful for the clinician to recall that MI is diagnosed when sensitive and specific biomarkers are elevated in the right clinical setting.¹ A characteristic rise and fall should be seen, as an initially elevated troponin may not result from ischemia.¹⁵ Troponin levels should be repeated at 6-hour intervals to define the clinical course.

Other Biomarkers

A number of novel cardiac biomarkers are being studied actively, including ischemia-modified albumin, high-sensitivity C-reactive protein (CRP), B-type natriuretic peptide (BNP), and others. In general, these markers detect conditions other than MI, and few currently have a well-recognized use in patients with ACS or those who are critically ill.

The theory behind ischemia-modified albumin is that ischemia changes the ability of the amino terminus of albumin to bind cobalt, and that this modified form can be measured in serum. Validation of this marker has been limited by lack of a gold standard for ischemia.¹⁵ Pregnancy-associated plasma protein-A is associated with neovascularization and is thus thought to be a potential marker for plaque rupture. Choline is released into the blood when phospholipids are cleaved, and thus might be a marker of ischemia and/or necrosis. None of these markers have been validated in the clinical setting, and none have been shown to add prognostic information to currently available techniques.

CRP is an acute-phase reactant synthesized in the liver and is a marker of inflammation. Levels of CRP have been used for detection and prevention of cardiac disease in ambulatory populations, and a recent study suggests that elevated circulating levels of CRP—measured using a high-sensitivity assay—may identify patients with normal low-density lipoprotein (LDL) levels who can benefit from therapy with statin.¹⁶ In critically ill patients, however, the value of measuring CRP is much less certain. Circulating concentrations of CRP may indicate the degree of inflammation, but how measurement of this analyte would impact management has not been defined in this context.

B-type natriuretic peptide is released by atrial and ventricular myocytes in response to increases in wall stress. BNP has been shown to facilitate the differential diagnosis of patients presenting with dyspnea, and to confer prognostic information in patients with heart failure.¹⁷

BNP is also released by ischemic myocardium. Circulating BNP levels are higher in patients with three-vessel coronary artery disease, tighter stenoses, and LAD disease. Higher BNP levels in ACS patients correlate with an increased risk of subsequent death, and BNP appears to confer information independent of other clinical markers. For example, in a study of 449 ACS patients, those with a high GRACE Risk Score and high serum BNP level were more likely to die than those with a high GRACE Risk Score and low serum BNP level.¹⁸ Interpretation of BNP levels can be complicated by the fact that women and older individuals have higher values, so age and gender-specific cutoffs may be needed. Obese individuals have lower values, but renal dysfunction increases BNP levels, sometimes dramatically. BNP levels also can be increased in the setting of right ventricular (RV) strain, including patients with pulmonary embolism, in whom both elevated circulating BNP and troponin levels predict worsened prognosis.¹⁹ BNP remains a good indicator of ventricular dysfunction myocardial wall stress, but what cutoff levels should be used in the ICU and what the clinician should do when serum BNP levels exceed those cutoff values remains unclear.

Conclusion

The ECG remains a valuable tool to diagnose myocardial injury. The clinician needs to remain aware of the myocardial injury imposters as well as methods to increase the sensitivity of the ECG in diagnosing myocardial injury. Biochemical markers, particularly troponin, are useful tools to confirm an ECG diagnosis as well as to predict prognosis in ACS patients and critically ill patients. In all settings, but especially in the ICU, these markers must be interpreted in the clinical context.