Exercise ECG testing is a common noninvasive diagnostic test used in the ED for low-to moderate-risk chest pain patients. It is easy to perform (49,50) and relatively safe after eliminating the possibility ofrest myocardial ischemia by serial 12-lead ECGs or myocardial necrosis by serial cardiac biomarkers (51). Compared with other noninvasive cardiac imaging, exercise ECG testing is relatively inexpensive (52). A meta-analysis of protocols including exercise ECG reported a sensitivity of 68% and a specificity of 77% for the diagnosis of myocardial ischemia (+LR 3.0; -LR 0.41), which is acceptable if applied to a low-risk patient population (53). Patients with negative exercise ECG tests are discharged home with outpatient follow-up within 1 wk, whereas patients with positive and nondiagnostic tests are admitted.
Exercise ECG testing has been incorporated into a number of chest pain protocols. The Cincinnati Heart ER strategy, established in 1991, admits low- to moderate-risk chest pain patients to a chest pain unit (13). Patients with ST-segment deviation on 12-lead ECG, a history of coronary artery disease (CAD), frequent or persistent symptoms typical of unstable angina, or hemodynamic instability are excluded from this protocol. The Cincinnati Heart ER strategy was initially a 9-h protocol using serial cardiac biomarkers, continuous ST-segment monitoring, and exercise ECG testing. Patients with negative testing are discharged home with close outpatient follow-up.
In a series of 2131 patients enrolled over a 6-yr period, 309 patients were admitted as a result of the Cincinnati Heart ER program (54). Thirty percent of the admitted patients were found to have AMI or unstable angina. Of the 1822 patients discharged home, 30-d follow-up was obtained in 1696 patients. Eight revascularizations (0.47%) and one death (0.06%) occurred at 30-d follow-up, indicating that this strategy was safe and effective (54). Over the past several years, this strategy has evolved into a 6-h protocol (Fig. 2), and continuous ST-segment monitoring has been eliminated. It has been replaced by single 12-lead ECGs obtained on presentation and at 3 and 6 h. Rest MPI is available as an alternative to graded exercise testing.
Similar protocols that have included patients with CAD have been developed at other institutions (55,56). Farkouh et al. (55) performed a randomized controlled study using a protocol that placed intermediate-risk unstable angina patients in a chest pain unit. Initial risk stratification was based on the Agency for Health Care Policy and Research
guidelines (Table 2) (57). Unlike the Cincinnati Heart ER strategy, 13.7% ofthe chest pain unit population had previous AMI. These patients had cardiac biomarkers measured at 2 and 4 h, along with ST-segment trend monitoring. If the biomarker results were negative, the patient underwent exercise ECG testing or nuclear stress testing. The investigators found that by using this protocol, hospitalizations were decreased by almost 50%, with no difference in event rates when compared to the strategy of hospital admission only (55).
Zalenski et al. (56) studied 317 patients with low-risk chest pain using Goldman's chest pain algorithm (58); these patients had a nondiagnostic ECG and had chest pain that was not typical for ACS. The protocol also included patients with a history of CAD. The protocol required serial biomarkers measured at 0, 4, 8, and 12 h. Ifthe biomarker results were negative, exercise ECG testing was performed. There were no adverse events as a result of exercise testing. The protocol was 88.8% sensitive and 47.7% specific (+LR 1.70; -LR 0.23) for AMI. For acute cardiac ischemia, the protocol was 90% sensitive and 50.5% specific (+LR 1.82; -LR 0.20). Acute cardiac ischemia was defined as positive stress echocardiogram or thallium, new ST-segment deviation, presence ofa dysrhythmia, cardiogenic shock, or cardiac arrest (56).
Immediate Exercise ECG Testing—UC Davis Protocol
Amsterdam et al. (59) evaluated a protocol implementing immediate exercise ECG testing without cardiac biomarker testing in 1000 low-risk chest pain patients (59). Patients
Agency for Health Care Policy and Research Risk Assessment for Unstable Angina
Any of the high or following features:
• History of prior MI or sudden death or other known history of CAD
• Definite angina: males >60 or females >70 yr of age
• Transient hemodynamic or ECG changes during pain
• Variant angina (pain with reversible ST-segment elevation)
• ST-segment elevation or depression >1 mm
• Marked symmetrical T-wave inversion in multiple precordial leads
Absence of high likelihood features and any of the following:
• Definite angina: males <60 or females <70 yr of age
• Probable angina: males >60 or females >70 yr of age
• Chest pain probably not angina in patients with diabetes
• Chest pain probably not angina and two or three risk factors other than diabetes
• Extracardiac vascular disease
• T-wave inversion >1 mm in leads with dominant R-waves
Absence of intermediate likelihood features but may have the following:
• Chest pain classified as probably not angina
• One risk factor other than diabetes
• T-wave flattening or inversion <1 mm in leads with dominant R waves
with previously documented CAD or Q-waves were included in the study. There were no adverse events as a result of exercise testing. Six hundred forty patients (64%) had negative exercise stress tests. Ofthe 582 discharged patients available for 30-d follow-up, one patient (0.17%) had a non-ST-elevation AMI, and another patient had a positive myocardial stress scintigraphy (0.17%). Nondiagnostic stress tests were seen in 235 patients (23.5%). No AMIs were diagnosed in these patients, but seven patients (3.0%) underwent revascularizations. One hundred twenty-five (12.5%) patients had positive exercise stress tests; ofthese, 4 patients had AMI (3.2%) and 12 patients had revascularizations (9.6%). The four patients with AMI had positive cardiac biomarkers obtained at presentation but received exercise ECG testing before the elevated cardiac biomarkers were known. As a result, the investigators recommended obtaining one set of cardiac biomarkers to maximize safety (59).
Rest MPI has been integrated into a number of chest pain protocols. MPI uses 99mtech-netium sestamibi and tetrofosmin, which are absorbed by viable myocardium by means of passive diffusion through the mitochondrial membrane. The amount of radionuclide uptake is proportional to coronary artery blood flow. Because its half-life is longer than 7 h and redistribution is limited, patients can be imaged several hours after injection. MPI can be performed immediately after the initial evaluation and with the tracer typically injected while the patient has pain at rest. If the patient is pain-free, the American Society ofNuclear Cardiology recommends that 99mtechnetium injections be given up to 2 h after the cessation of pain (60). Rest MPI is ideal for patients at low to moderate risk of ACS.
Compared with initial serum cardiac biomarkers, rest MPI is more sensitive for detecting AMI. In one study, rest MPI was 92% sensitive whereas the initial cTnl was only 39% sensitive in low- to moderate-risk patients with chest pain (26). The additional benefit ofrest MPI is the detection ofmyocardial ischemia before necrosis is present (26). Rest MPI also predicts 30-d AMI and revascularization with a reported sensitivity of82% and a specificity of 83% (+LR4.81; -LR 0.22) (46). In addition to being cost-effective, this diagnostic tool improves physician decision making, reducing hospital admissions up to 10% (61).
Serial cardiac biomarker testing and rest MPI offer complementary information. Kontos et al. (62) found that only 76% of patients with positive serial cTnI had a positive myocardial perfusion image. In a study of620 low- to moderate-risk chest pain patients, 5 patients with AMI had a negative rest MPI. However, four ofthese five patients had positive serial cTnI measurements (26). Patients with positive serial cTnI but a negative rest MPI tended to have a higher prevalence of insignificant CAD and a smaller infarct size (62). This discrepancy may exist because a minimum of 3-5% myocardial involvement is required to be detected by MPI (63).
Rest MPI has some advantages over the other aforementioned methods. The test is performed at rest and is not dependent on the patient's ability to ambulate. Rest MPI also has the ability to detect unstable angina (i.e., ischemia in the absence ofnecrosis) and may provide earlier diagnosis of AMI compared with serial cardiac biomarker testing. However, there are limitations to rest MPI. The specificity has been reported to be 67% and, therefore, there are a relatively high number of false positives (26). Because rest MPI is reflective of myocardial tissue viability, patients with previous AMI will have an abnormal rest MPI. For this reason, this patient population is excluded from chest pain protocols implementing rest MPI, and these patients require other testing to confirm acute myocardial ischemia or AMI.
The Medical College of Virginia Hospitals ofthe Virginia Commonwealth University have developed a comprehensive strategy incorporating MPI using a five-level approach to the evaluation of chest pain (Fig. 3) (46). Patients categorized in levels 1 and 2 are considered to be at high risk of AMI or unstable angina and are admitted to an intensive care unit for appropriate treatment. Level 5 patients are considered to be at very low risk of AMI or unstable angina and are discharged without any further cardiac testing. Patients assigned to level 3 have typical symptoms lasting >30 min but have nondiagnostic ECGs and no history of CAD. Level 3 patients are considered to be at moderate risk of AMI or unstable angina. These patients undergo immediate MPI performed in conjunction with serial ECGs and cardiac biomarkers performed at 0, 3, 6, and 8 h. Ifthe MPI and serial cardiac biomarkers are both negative, level 3 patients are referred for stress imaging within 12-24 h. Level 4 patients have typical symptoms lasting <30 min or prolonged atypical chest pain. The ECG is nondiagnostic, and a history of CAD is absent. These patients are considered to be at low to moderate risk and receive immediate MPI without serial cardiac biomarker testing. If results are negative, these patients are discharged without serial cardiac biomarkers and with a follow-up evaluation scheduled within 48 h. Any patient with a positive MPI or positive cardiac biomarkers is treated and becomes retriaged to the level 2 category.
Using this strategy, Tatum et al. (46) enrolled 1187 consecutive patients. Among level 3 patients, 3% were diagnosed with AMI and 17% underwent index revascularization.
Among patients categorized as level 4, only 0.7% had AMI and 2.5% received index revascularization. Of note, all patients diagnosed with AMI had abnormal MPIs. Overall, the chest pain protocol was 92% sensitive and 100% specific (+LR >100; -LR 0.08) for AMI. This strategy also appeared to be cost-effective. Although the cost for individual ED visits increased, it was offset by a decrease in hospital admissions, shorter length of stay, and a more effective use of coronary angiography, thus decreasing overall costs (45,46).
Was this article helpful?