The Evolution of Chest Pain Centers:
The Impact of Technological Advances on The Evaluation of Patients With Possible Acute Ischemic Coronary Syndrome

The accurate diagnosis and management of patients with chest pain is quite challenging. Approximately 4 million patients are admitted across the United States each year for possible acute ischemic coronary syndrome (AICS). Of these patients, about 300,000 will suffer sudden death, and 900,000 will be diagnosed with an acute myocardial infarction (AMI). Another 900,000 patients will be diagnosed with angina or unstable angina. More than 2 million patients will be found to have other causes for their chest pain. The goal for the clinician is to admit the patient who truly has a life-threatening disease while avoiding the inappropriate release of patients who have AMI because of the potential for these individuals to suffer morbidity and mortality. This must be accomplished while decreasing the overall number of patients admitted with a noncardiac cause of their symptoms. Chest pain centers have evolved over the last 15 years that can accomplish these goals.

Cost-effective care for the chest pain patient is important to all emergency care providers for the following reasons:
(1) chest pain is a common patient complaint, accounting for 2% to 8% of emergency department (ED) visits;
(2) the presentation is rarely "typical," especially in the elderly and in patients with diabetes mellitus; (3) the 12-lead electrocardiogram (ECG) is diagnostic in only half of the cases of AMI. (1)

Finally, the risk of misdiagnosis in patients with possible AICS is substantial. About 1.5% to 5% of patients with AMI are released from the ED across the United States. (2) Some of these patients develop ventricular fibrillation or pump failure hours or days after a physician's diagnosis of noncardiac chest pain, and they often have a poor outcome. Therefore, about 20% of the malpractice dollars awarded in emergency medical care are related to the emergent care of patients with AICS. (3) Most physicians order a 12-lead ECG after a thorough history and physical examination; if available, a single total serum creatine kinase (CK) or CK-MB will be obtained.

A number of technological advances in the past decade have permitted a more rapid approach to the accurate diagnosis of AMI than the time-honored 3-day rule-out AMI admission.

New Technology for Serum Cardiac Markers

The development of immunochemical testing for serum cardiac markers is a key technologic development. Prior to the late 1980s, diagnostic testing for cardiac enzymes was quite time consuming, often occurring only once each day in most hospital laboratories. New immunochemical techniques for measuring the mass rather than the activity of an enzyme permit detection of smaller amounts of the protein markers and, therefore, earlier diagnosis is possible. In addition, new laboratory techniques now permit rapid measurement of a wide variety of markers for myocardial cell necrosis. Several serum proteins have been evaluated to determine sensitivity and specificity for AMI, including CK-MB mass measurements, (4,5) CK-MB isoforms, (6) myoglobin, (7,8) myosin light chains, (9) troponin I, (10,11) and troponin T. (12,13 ) Successful use of these tests is accomplished through serial sampling. A single sample of any of the currently available markers is insensitive at the time of the patient's admission to the ED.

Of the various serum markers for AMI, measurement of cardiac troponin T (cTnT) is as sensitive and more specific than the traditional CK-MB test. (14) In particular, cTnT testing appears to be more sensitive than CK-MB for detecting small areas of myocardial necrosis. (13,15) After the onset of AMI, troponin T is released from deteriorating myocardial cells, appearing in the serum within 2 to 6 hours. (12,16,17) Because patients typically wait at home for a variable period of time prior to ED presentation, approximately 50% of patients who are ultimately found to have AMI will have a positive troponin T assay on admission, which increases to 80% after 2 to 3 hours in the ED. (18,19)

Quick Overview

Serial 12-Lead Electrocardiography

Another interesting technological improvement has been the development of serial ECGs or continuous ST-segment trend monitoring. (20-22) The ST-segment trend monitor obtains a new 12-lead ECG every 20 seconds and compares that ECG with the others in its memory bank. (23) If there is a significant change in the ST segment, typically defined as 1 mm or greater ST-segment elevation or depression, an alarm is sounded. This machine may permit the detection of silent ischemia. It may also permit the clinician in the ED to identify candidates for acute intervention for AMI who otherwise would have been missed. (24-26)

Echocardiography

Echocardiography has been used in a variety of settings to enhance the diagnosis of AICS. (27) Sabia and colleagues from the University of Virginia found that 27 of 29 AMI patients had regional wall motion abnormalities on an ED echocardiogram performed within 4 hours of the onset of chest pain. (28) All patients who subsequently developed cardiogenic shock, life-threatening arrhythmias, or post-MI angina had regional wall motion abnormalities on this initial ED echocardiogram. This study was limited, however, by a small number of patients with AMI and the systematic exclusion of patients who had pain for greater than 4 hours.

Echocardiography was performed on 43 patients in the ED by Peels et al. (29) All patients had selective coronary angiography within 3 weeks of ED presentation. Regional wall motion abnormalities were present in the emergency setting in 26 of the 43 patients. Twenty-two of the patients with regional wall motion abnormalities were found to have coronary artery disease (CAD). Three other patients with CAD did not have regional wall motion abnormalities on their ED echocardiograms. This study was limited by including only a select group of patients at high risk for CAD who were thought to need a cardiac catheterization by the treating physicians.

The timing of the echocardiogram is crucial. In a study by Levitt et al, echocardiograms were performed within 12 hours of ED presentation. (30) They found that the echocardiogram added little to the accuracy of AMI diagnosis after the results of CK-MB were known. A recent study documents that patients with unstable angina can have wall motion abnormalities that resolve in a variable period of time, which ranges from less than 2 hours to more than 24 hours. (31) It is likely that some of the patients in Levitt's study who ruled out for AMI yet had wall motion abnormalities on echocardiography had unstable angina. In such cases the echocardiogram should not be viewed as falsely positive for AMI but rather as truly positive for AICS. Other patients with small nonp;Q-wave MIs may not have wall motion abnormalities detectable by echocardiography. The data of Sabia et al would suggest the possibility that those patients may be at low risk for a bad outcome.

Current evidence indicates that echocardiography is only helpful when it is performed early in the evaluation of the chest pain patient. Our clinical experience with over 1,350 chest pain patients who have entered our Heart ER chest pain evaluation and treatment unit confirms the conclusion of Levitt et al that an echocardiogram performed hours after ED arrival rarely is helpful. (32) Additional study is needed to learn if the echocardiogram should only be performed early after ED presentation, or only for a certain subset of patients with possible ACIS.

Graded Exercise Testing

Excluding fixed coronary artery stenosis without cardiac catheterization can be quite challenging as well. It is well known that graded exercise testing has some value in the diagnosis of angina, but it suffers from poor sensitivity in some patients and a high rate of false-positive tests, particularly in women. Mark et al used the results of treadmill testing in an outpatient population to successfully predict outcomes. (33) Exercise testing has been performed in the ED as well. In two small case series, it was demonstrated that exercise testing can be performed safely in the ED. (34,35) Our own larger series of patients demonstrates that patients can run safely in the ED. (32) Other centers have evaluated exercise testing early after hospital admission and then successfully extended this diagnostic modality into the emergency setting.(36,37)

Radionuclide Testing

The use of radionuclide testing in patients with possible AICS presenting to the ED represents an important new area of research. Radionuclide testing with technetium-99m sestamibi, a new agent that persists in the myocardium after uptake for multiple hours after injection without redistribution, promises to improve the functional evaluation of patients with possible AICS in the emergency setting. (38-42)

Development of Chest Pain Centers

In 1981, Bahr noted that patients with chest pain were not evaluated expeditiously in the ED. He established the first chest pain center with the idea of speeding the evaluation and treatment of patients with AMI. (43) Since that time, the concept has expanded to include the quick evaluation of all chest pain patients. (44-46) A recent editorial predicted that there will be 3,000 hospitals with chest pain centers in the United States in the year 2000. (47) Establishment of a chest pain center is a complicated process. Support of emergency medicine, cardiology, nursing, and hospital administration, and cardiology is critically important.

Quick Overview

University of Cincinnati Chest Pain Protocol

Patients greater than 25 years old having symptoms consistent with low to intermediate likelihood of unstable angina by Braunwald's criteria are eligible for evaluation in our Heart ER program. (48) Patients with symptoms suggestive of a high likelihood of unstable angina, or having ST-segment changes on ECG are given appropriate therapy and admitted. (32)

Serum cardiac enzyme markers are drawn at 0, 3, 6, and 9 hours after entry. The patient is also evaluated by a serial 12-lead ST-segment monitor. The machine is set to sound an alarm if there is any positive or negative change greater than or equal to 1 mm in the ST segment that persists for at least 1 minute.

At the end of the 9-hour diagnostic assessment with serial CK-MB levels and ST-segment trend monitoring, a cardiologist evaluates the patient. During the first 4 years of the Heart ER program, all patients were to have a diagnostic echocardiogram. Our data suggest that the echocardiography study adds little value for the vast majority of patients when performed at the end of the 9-hour protocol. The other studies by Sabia and Peels suggest that echocardiography may have value when used earlier in the patient's evaluation, so the cardiologist now performs an echocardiogram if it will provide useful diagnostic information for a particular patient. After the completion of the 9-hour protocol, the cardiologist decides on the appropriate test to exclude the possibility of a fixed coronary artery lesion. During the first 4 years of the protocol, patients were to have the graded exercise test using a maximal Bruce protocol whenever possible. Now the cardiologist performs a risk
assessment and determines the best diagnostic test for a particular patient. Available choices include cardiac catheterization, stress echocardiography, outpatient stress or pharmacologic thallium nuclear study, or standard exercise electrocardiographic testing, which is available at all times in the ED. This protocol improvement will permit the evaluation of patients who cannot exercise, such as those with claudication and chronic obstructive pulmonary disease, or patients with baseline ECG abnormalities precluding a standard graded exercise test.

University of Cincinnati Results

Our center evaluated 1,356 patients from October 17, 1991, through June 30, 1995. One hundred and ninety seven of those patients were admitted to the hospital. The mean age of the patients was 44.9 years with a range of 19 through 89. Eighty-five percent of the patients had an echocardiogram, and 79% received a graded exercise test.

Seventeen of the 1,356 patients had positive CK-MB enzyme tests. Twelve (1%) had AMI. Two likely had small myocardial infarcts that were misdiagnosed as unstable angina. Three had spurious elevation of CK-MB. Sixty-four of the 197 admitted patients had discharge diagnoses demonstrating cardiac disease. There were 12 AMIs with 7 patients who received percutaneous transluminal coronary angioplasty (PTCA) and one who had coronary artery bypass surgery (CABG). Thirty-eight other patients had angina or unstable angina, with 16 patients undergoing PTCA and 4 who required CABG. Eight admitted patients had cocaine-related chest pain, and 6 had other nonischemic cardiac discharge diagnoses.

Of 1,356 patients, 149 admitted to cocaine use; 17 of these patients were admitted. Four were admitted because of ST-segment changes on the ST-segment trend monitor, while 3 had ongoing chest pain suggesting coronary artery spasm. Two had elevated CK-MB and total CK but a normal Relative Index (mass CK-MB/activity of total CK). One patient had a large stroke. (49)

This protocol appears to be less expensive than hospital admission. (50) We analyzed costs in a group of chest pain center patients and compared them with a group of hospitalized patients who ruled out for AMI and were discharged in 3 days or less. The mean hospital cost for the chest pain center group was $995 ± $352 compared with $2,026 ± $841 in the hospitalized patients. (51)

Quick Overview