ABSTRACT
Approximately 6 million people annually undergo evaluation in the emergency department (ED) for acute chest pain, at a cost of several billion dollars in the United States.1 Most of these patients are admitted to the hospital, and their average length of stay is 1.9 days. Nearly half of hospitalized patients with unstable angina eventually receive a non-cardiac-related diagnosis2-most frequently panic attack,3 gastroesophageal reflux, or musculoskeletal causes.2 The low-risk patients neither require nor benefit from management in a coronary care unit (CCU). In addition, unnecessary admission to a CCU is inordinately costly, over $2000 per day, and it also reduces the availability of vital CCU beds. Finally, unnecessary CCU admission imposes both undue stress and potential morbidity. Although this low threshold for CCU admission results in a high ratio (4:1 to 5:1) of patients with nonischemic causes of chest pain to the number admitted with an ischemic event, 5% of patients with myocardial infarction are inappropriately discharged from the
ED.4 In the United States, an estimated 25% of lawsuits concerning emergency care involve errors in the diagnosis of myocardial infarction.5 In response to this need for improved emergency care, chest pain centers have been established, in which patients are monitored and observed for 24 h. This observation protocol usually incorporates selective history, physical examination, and electrocardiographic variables, with the concomitant use of serum markers of myocardial cell death (usually troponins). This primary risk stratification has limitations. Clinical variables have a low specificity, the electrocardiogram (ECG) has a high diagnostic accuracy in myocardial infarction but a low accuracy in unstable angina, and biochemical markers of cardiac damage are relatively late findings.1,2 In this setting, the incremental value of imaging and stress-test techniques has been documented,6-9 although the dramatic increase in resource use and intensity of cardiac imaging application does not seem to be matched by increased quality of care.10
