Biomarkers in the emergency workup of chest pain: Uses, limitations, and future

Each year in the United States, more than 8 million people come to the emergency department with chest pain, but only a minority are eventually diagnosed with a heart attack.¹

Confronted with signs and symptoms that could represent an acute coronary syndrome, clinicians need to know whether the patient has a benign condition and can safely be sent home or is in urgent need of hospitalization—and they need to do so in a safe, timely, and cost-effective manner.^2,3

Testing for biomarkers of cardiac injury, especially troponins I and T, is an accepted part of the assessment of chest pain. However, the interpretation of these cardiac biomarkers is complicated by the fact they can be elevated from noncoronary causes of chest pain such as pulmonary embolism or renal impairment, and thus should be considered only as part of the patient’s total clinical picture. This uncertainty can result in longer hospital stays and increased testing.

Thus, researchers are searching for new biomarkers that could allow for more rapid and accurate diagnosis and estimation of prognosis.

In this article we will examine the advantages and limitations of measuring cardiac biomarkers. We then discuss the emerging data on new biomarkers, including the very promising high-sensitivity troponin assays, cystatin C, and other markers, and the potential for biomarkers to be used instead of or in combination with stress testing in the evaluation of patients who have no initial evidence of ischemia.

SCENARIO 1: ELEVATED TROPONIN AND ST-SEGMENT ELEVATION

A 46-year-old woman presents to the emergency department with chest pain that started 2 hours earlier. Electrocardiography (ECG) initially shows sinus tachycardia with ST-segment depression and negative T waves in lead aVL. Her cardiac biomarker values (troponin I and creatine kinase MB) are normal. Repeated troponin I measurements show elevations of 250 ng/L, whereas her creatine kinase MB level is within the optimal range. Coronary angiography is unremarkable. Echocardiography shows right ventricular pressure overload in the pulmonary artery and the right ventricle. How should this patient be further evaluated?

SCENARIO 2: ELEVATED TROPONIN AND LEFT VENTRICULAR HYPERTROPHY

A 47-year-old man is admitted with worsening dyspnea and chest pain that worsens with coughing and inspiration. He has a history of end-stage renal disease secondary to poorly controlled hypertension and is being treated with hemodialysis, which he missed for the past 4 weeks while failing to take his hypertension medication. His blood pressure is 270/130 mm Hg. Chest auscultation reveals signs of pulmonary edema—ie, crackles at the end of inspiration. His troponin T level is 394 ng/L. ECG indicates left ventricular hypertrophy. How should this patient be further evaluated?

TROPONIN IS SPECIFIC FOR INJURY, BUT NOT FOR INFARCTION

American College of Cardiology and American Heart Association (ACC/AHA) guidelines⁴ recommend that clinicians ask themselves two questions: what is the likelihood that the patient is truly having an acute coronary syndrome secondary to coronary artery disease, and what is the likelihood of an adverse clinical outcome? Clues come from the initial measurements of biomarkers of cardiac injury, history, physical examination, and ECG (Table 1),⁵ and subsequent care is based on the estimated degree of risk.

Troponin revolutionized the diagnosis and risk stratification of chest pain. The ACC/AHA guidelines call for measuring biomarkers—preferably troponin—in all patients who present with chest discomfort consistent with an acute coronary syndrome.^4,6

Cardiac troponins I and T have been the biomarkers of choice for detecting myocardial injury,^4,6 since elevated concentrations are highly sensitive and tissue-specific.⁷ Moreover, they identify patients at short-term and long-term risk of cardiac events.^4,8

The introduction of troponin testing led to a substantial increase in the rate of diagnosis of myocardial infarction (MI), with an increase in cardiac care unit admissions of more than 20%.^9,10 This was partly because troponin is released into the blood with even minute myocardial damage, so that some patients who previously would have been diagnosed with unstable angina are now found to have non-ST-segment-elevation MI.¹⁰ However, the increase in admissions may also represent an increase in misdiagnoses, with many clinicians equating an elevated troponin level with acute MI.¹¹

Although an elevated troponin level is 100% specific for myocardial injury, it is not synonymous with MI.¹² Myocardial injury can be caused by a cardiac condition such as tachyarrhythmia, cardiac trauma, congestive heart failure, ventricular hypertrophy, myocarditis, or pericarditis, or by a noncardiac condition such as sepsis, respiratory failure, pulmonary embolism, pulmonary hypertension, cancer chemotherapy, or renal insufficiency.^4,13 Therefore, to avoid a misdiagnosis of MI, the troponin level must be considered in the clinical context.

In fact, Alcalai et al¹¹ noted that almost half of patients with elevated troponin did not really have an acute coronary syndrome. More importantly, in-hospital and long-term survival rates were significantly better for patients with an acute coronary syndrome than for those without, illustrating the importance of identifying and treating the true disease instead of mislabeling the problem as MI.

Bayesian theory predicts that patients with chest pain who have elevated troponin are less likely to truly have an acute coronary syndrome if the rest of their clinical presentation indicates a low probability for heart disease.¹⁴ Indeed, when McDonald et al¹⁵ used a risk-scoring index based on sex, a history of heart failure or coronary artery disease, the ECG, and use of aspirin, the positive predictive value of an abnormal troponin level was 83% at a risk score of 4 or greater, 63% at a score of 3, 52% at a score of 2, 32% at a score of 1, and 29% at a score of 0.

Thus, cardiac biomarkers are not a substitute for traditional clinical assessment, but rather should be used “in conjunction with the clinical history, physical examination, and interpretation of the ECG.”⁶ Consequently, diagnostic protocols that incorporate pretest clinical features to identify low-risk patients have a higher negative predictive value.

This was illustrated in a study by Than et al¹⁶ that aimed to prospectively validate the safety of an accelerated diagnostic protocol to assess chest pain suggestive of an acute coronary syndrome. The protocol included a structured pretest probability scoring method (ie, the Thrombolysis in Myocardial Infarction [TIMI] score), ECG, and a point-of-care biomarker panel of troponin, creatine kinase MB, and myoglobin. The protocol had a negative predictive value of 99.1%, whereas the use of biomarkers alone had a value of 96.1%.