INPATIENT ADMISSION In settings where extended observation and definitive diagnostic testing is not available in the ED, all patients whose presentation suggests any reasonable plausibility of an acute ischemic event must be admitted to an inpatient bed. Even in the presence of an ED-based cardiac evaluation protocol, inpatient admission is often prudent for patients whose presentation suggests a low probability of ischemia but who also have other acute medical, surgical, or psychiatric diseases requiring a prolonged or intensive level of attention, which cannot be optimally given in the ED. Once the need for inpatient admission has been determined, further stratification based on assessment of the patient's short-term risk of morbidity or mortality can be made based on the patient's history and physical exam, initial ECG,9 and early myocardial marker measurement.12,2 22 Patients with a prior history of coronary artery disease, evidence of congestive heart failure on physical exam, recurrent chest pain in the ED, or new or presumed new ischemic ECG changes are at higher short-term risk and may be more appropriately managed in an intermediate-care (step-down) unit.6 Conversely, patients whose initial ED ECG is normal or unchanged from a previous ECG have a very low risk of adverse events and can safely be evaluated on a monitored floor or telemetry bed. Those with nonspecific changes on the initial ECG represent an intermediate risk group. A single myocardial marker measurement soon after ED presentation can also identify those patients at greater risk from among those with atypical presentations. 22
ED OBSERVATION/MONITORING In 1984, the Multicenter Chest Pain Study Group was formed to develop new lower-cost strategies for managing "ROMI" (rule out MI) patients. In 1991, this group published the results of a seven-center investigation of 2684 patients admitted for chest pain and reported that the diagnosis of infarction could have been safely excluded within a 12-h observation period among a subgroup identifiable at presentation as having a low probability of AMI. 28 The authors also suggest routine predischarge stress testing of these patients in order to reduce the risk of discharging patients with unstable coronary syndromes prematurely.
This approach was further refined by Gibler et al. in a study of 1010 patients admitted to an ED-based chest pain evaluation and treatment unit ("Heart ER") over 32 months. Patients with symptoms consistent but not highly suggestive of acute ischemia were observed for 9 h with continuous 12-lead ST-segment ECG monitoring and serial CK-MB testing at 0, 3, 6, and 9 h after presentation. Those who completed a negative 9-h evaluation subsequently underwent echocardiography followed by graded exercise stress testing in the ED prior to discharge. Utilizing this approach, 82.1 percent of patients were released home from the cardiac evaluation unit (CEU).29
The ideal length of observation (12, 9, or 8 h), the best choice and timing of myocardial marker measurements, and the value of continuous ST-segment monitoring are among the many questions that have not yet been answered definitively. Additionally, while several studies have documented cost savings associated with use of an ED-based CEU compared with traditional inpatient admission, the cost effectiveness of their widespread use is still not uniformly accepted.
Although AMI can be reliably excluded by this type of observation protocol ( Fig... ...45-4), normal serial ECGs and myocardial marker measurements do not preclude the presence of other ACSs (i.e., unstable angina), which may still put the patient at high risk for a subsequent adverse event. Therefore, further evaluation is usually indicated prior to discharge. The various forms of stress testing (with or without myocardial imaging) currently offer the best noninvasive method to both predict the presence of coronary artery disease as well as assess prognosis. In the Cincinnati study previously mentioned, 29 791 patients underwent treadmill exercise stress testing after a negative observation period with no adverse events reported during testing. Among these, 9 of the patients had positive tests and 4 of these subsequently confirmed to have significant coronary artery disease not otherwise identified (positive predictive value = 44.4 percent). Of the 782 negative results, 10 were false negatives, yielding a negative predictive value of 98.7 percent. The authors concluded that although a negative observation period and stress test still does not completely rule out the possibility of coronary artery disease, they do identify a large group of patients who could be safely released from the ED with subsequent follow-up.29 In a subsequent study, 86 percent of 502 patients who underwent a similar observation protocol and ECG stress testing were discharged after a negative evaluation, with no deaths or AMIs reported after 5 months of follow-up. In this investigation, 24 patients with subsequently confirmed ischemic heart disease were identified only through positive ED stress tests. However, the effectiveness of this technology is still not uniformly accepted. A recent consensus panel report from the National Heart Attack Alert Program cited the need for additional investigation and concluded that "ECG exercise stress testing in the ED cannot be recommended in the absence of additional data demonstrating safety and effectiveness."26
FIG. 45-4. Algorithm for the evaluation of low-risk patients.
IMMEDIATE SCREENING In addition to admission or ED-based observation, immediate myocardial perfusion imaging, echocardiography, or stress testing without prior observation have been suggested as rapid, safe, and cost-effective alternatives among selected patients. However, few ED-based studies have been published concerning these approaches and many questions remain unanswered concerning their limitations and relative value.
Because of its minimal redistribution, myocardial perfusion imaging using 99mTc sestamibi (sestamibi scanning) is particularly suited for the early ED assessment of both diagnosis and prognosis among patients with ongoing chest pain. The agent can be injected in the ED upon presentation without interfering with other therapeutic or diagnostic interventions. Imaging can then be delayed for up to 3 h without significant degradation of the resulting image. With this technique, the image acquired will represent the state of myocardial perfusion at the time of injection, and a normal perfusion scan should therefore reliably exclude myocardial ischemia as a cause of the patient's chest pain. Indeed, multiple investigations have confirmed that an abnormal study has a high sensitivity (94 to 100 percent) for the prediction of AMI when injection occurred during active pain.
The value of resting sestamibi scanning in patients who present after the resolution of chest pain is less clear. Some investigators have reported a substantial decrease in diagnostic accuracy related to increasing delay from symptom resolution to tracer injection. However, in another study among those with greater than 30 min of chest pain within the 12 h prior to presentation, sensitivity for AMI remained 100 percent. 30 In the largest published series to date, 100 percent sensitivity for AMI was also reported among 438 low-risk patients enrolled and scanned without regard to the presence or absence of chest pain at the time of injection. 31 In this study, among the 338 "low-risk" patients with normal perfusion scans in the ED, there were no AMIs or deaths within the 1-year follow-up period. However, seven (2.1 percent) low-risk patients with normal scans did require a coronary revascularization procedure within 1 month of their ED visit. Taken as a whole, the current literature suggests that low-risk patients with negative sestamibi scans may be safely discharged from the ED without an extended observation period. Pharmacologic or exercise stress imaging using sestamibi may subsequently be performed either after an observation period or during a follow-up visit to assess for the presence of stress-induced ischemia.
Several small studies have been published in which chest pain patients were assessed with standard echocardiography soon after ED presentation. Among these, the sensitivity for the detection of acute ischemia or AMI among all patients ranged from 88 to 96 percent, significantly higher than that of the 12-lead ECG. However, even among selected patients with adequate studies who are evaluated during ongoing chest pain, false-negative echocardiograms may occur in up to 8 percent of AMI patients, thus precluding the possibility of discharging a patient on the basis of a negative study. Primarily due to this limitation, the National Heart Attack Alert Program Working Group on echocardiography has concluded that this tool cannot currently be recommended for ED use. 32
A few reports have advocated immediate exercise testing of low-risk chest pain patients presenting to the ED without prior observation or ruling out of AMI. One small series of 32 patients who presented with "chest pain suggestive of cardiac origin but not typical of angina," normal or nondiagnostic ECG, and less than two cardiac risk factors reported that all stress tests were negative and no adverse events occurred. Another investigation of 212 ED patients with similar inclusion criteria found that 13 percent had a positive stress test, 59 percent a negative test, and 2 percent a nondiagnostic result with no adverse events in any group. 33 The limited amount of available data concerning both clinical effectiveness and safety preclude routine application of this approach at the present time.
Was this article helpful?