Original Article
Ruling out Acute Myocardial Infarction — A Prospective Multicenter Validation of a 12-Hour Strategy for Patients at Low Risk
N Engl J Med 1991; 324:1239-1246May 2, 1991
- Abstract
- Abstract
Background.
Although previous investigations have suggested that 24 hours is required to exclude acute myocardial infarction in patients who are admitted to a coronary care unit for the evaluation of acute chest pain, we hypothesized that a 12-hour period might be adequate for patients with a low probability of infarction at the time of admission.
Methods.
Using a Bayesian model, we developed a strategy to identify candidates for a shorter period of observation from an analysis of a derivation set of 976 patients with acute chest pain who were admitted to three teaching and four community hospitals. In the derivation set, patients whose clinical characteristics in the emergency room predicted a low (≤7 percent) probability of myocardial infarction had only a 0.4 percent risk of infarction if they had neither abnormal levels of cardiac enzymes nor recurrent ischemic pain during the first 12 hours of hospitalization. In an independent testing set of 2684 patients from the seven hospitals, 957 admitted patients (36 percent) were classified as candidates for this 12-hour period of observation according to a previously published multivariate algorithm. Few of these patients were actually transferred from a monitored setting at 12 hours.
Results.
Of the 771 candidates for a 12-hour period of observation who did not have enzyme abnormalities or recurrent pain during the first 12 hours, 4 (0.5 percent) were subsequently found to have acute myocardial infarction, and only 3 (0.4 percent) died after primary cardiac arrests, all of which occurred three to five days after admission. Rates of other major cardiovascular complications were low in the patients who might have been transferred from the coronary care unit after 12 hours with this strategy. In patients with a higher initial risk of infarction, the standard strategy of 24-hour observation identified all but 11 of 739 acute myocardial infarctions (1 percent).
Conclusions.
Emergency room clinical data can be used to identify a large subgroup of patients for whom a 12-hour period of observation is normally sufficient to exclude acute myocardial infarction. Patient-specific evaluation and treatment can then proceed without the restrictions imposed by "rule-out" protocols for myocardial infarction. (N Engl J Med 1991; 324:1239–46.)
Media in This Article
Figure 1
Prospectively Validated Multivariate Algorithm for the Prediction of a Patient's Risk of Acute Myocardial Infarction on the Basis of Emergency Room Data.Figure 2
Predicting the Probability of Acute Myocardial Infarction If Cardiac Enzyme Levels Are Normal and Ischemic Chest Pain Does Not Recur after 12 (Open Squares) and 24 (Solid Squares) Hours of Observation in a Monitored Setting.Article Activity
- Article
IN recent years, economic pressure on the health care system has intensified interest in using risk stratification to increase efficiency in the care of patients with suspected acute myocardial infarction.1 To avoid the initial triage to coronary care units of patients who are unlikely to benefit from them, some investigations have focused on identifying emergency room patients with a low risk of acute myocardial infarction,2 3 4 5 6 acute ischemic heart disease,7 or life-threatening complications.8 9 10 Data indicate, however, that considerable and perhaps greater economic gains can be achieved by identifying patients after admission who are at sufficiently low risk of acute myocardial infarction and its complications that they can safely be transferred to beds outside the coronary care unit.11 12 13
Previous investigations have found that patients at low risk of infarction and its complications can be identified 24 hours after admission by the absence of clinical complications and evidence of acute myocardial infarction,14 , 15 but that a 12-hour period of observation is not sufficient to recognize acute myocardial infarction in as many as 23 percent of patients who are ultimately given this diagnosis.14 Although these data suggest that 24 hours is the minimal period of observation needed to exclude infarction safely in patients with acute chest pain, we hypothesized that a 12-hour period might be appropriate for patients whose probability of acute myocardial infarction at the time of admission was low, although not low enough for them to be discharged. We therefore developed and prospectively tested a strategy for identifying a large subgroup of patients admitted for the evaluation of acute chest pain who would be candidates for a shorter observation period.
Methods
Patient Population
The Multicenter Chest Pain Study (see Appendix) is a cooperative investigation of the clinical characteristics and outcomes of emergency room patients with acute chest pain.2 3 4 5 , 10 , 14 , 16 17 18 19 20 21 22 23 All patients who were 30 years or older who went to the emergency room with a report of anterior, precordial, or left-sided chest pain unexplained by obvious local trauma or abnormalities on a chest roentgenogram were eligible for the study for up to three visits. The participating institutions were Brigham and Women's Hospital, Boston (which began enrolling patients in January 1984); Yale–New Haven Hospital, New Haven, Connecticut (December 1983); Danbury Hospital, Danbury, Connecticut (January 1984); Milford Hospital, Milford, Connecticut (June 1984); St. Mary's Hospital, Waterbury, Connecticut (April 1984); University of Cincinnati Hospital, Cincinnati (July 1984); and William Beaumont Hospital, Royal Oak, Michigan (June 1985).
Clinical data from the emergency room evaluation, including history, results of physical examination, the presence or absence of an earlier electrocardiogram, and interpretation of the emergency room electrocardiogram, were recorded as part of a detailed protocol by the physician in the emergency room or by a research nurse. The person who recorded these data had no knowledge of the patient's course after treatment in the emergency room and thus could not be influenced by it.
This report includes data on 3660 (30 percent) of the 12,140 patients who were enrolled in the Chest Pain Study from December 1983 to November 1986. This subgroup included all those who were admitted to the hospital; who underwent at least two sets of cardiac-enzyme tests (for creatine kinase and its isoenzymes) during the first 12 hours after admission, or met the study criteria for acute myocardial infarction during this period through sudden cardiac death without other known cause (see below); for whom sufficient emergency room clinical data were available to allow us to estimate the risk of acute myocardial infarction at the time of admission using a previously published multivariate algorithm2; and for whom, if ischemic chest pain recurred during the first two days of hospitalization, there were data on the time of day at which the pain occurred.
Definition of Outcomes
Detailed information on the definition of outcomes in the Chest Pain Study have previously been published.2 In the patients in this report, acute myocardial infarction was diagnosed if there was a characteristic evolution in serum enzyme levels, as evidenced by detection of (1) the MB isoenzyme of creatine kinase in more than trace amounts by the qualitative electrophoretic assay or in amounts of at least 5 percent of the total creatine kinase level, with a typical rise and fall, by quantitative assays; (2) a typical rise and fall of the total creatine kinase level, with a peak value at least twice the upper limit of normal, if creatine kinase MB was not assayed; or (3) a level of lactate dehydrogenase isoenzyme 1 higher than that of isoenzyme 2 in the absence of hemolysis or renal infarction. Patients who had sudden cardiac arrest in the hospital were also classified as having acute myocardial infarction if they died before enzymatic confirmation of myocardial necrosis could be obtained and there was no other explanation for the arrest. In all such patients, the presenting characteristics, electrocardiograms, and clinical courses were most consistent with acute myocardial infarction. Of the 1122 patients who met the criteria for acute myocardial infarction in this report, 1005 (90 percent) were given the diagnosis on the basis of elevated creatine kinase MB levels in the setting of elevated total creatine kinase levels.
Unstable angina was diagnosed if the patient's original emergency room syndrome of chest pain was either new or worse (in frequency, severity, or duration) than his or her chronic anginal syndrome, and if the diagnosis of angina was made by the senior clinician involved in the case.
Derivation of Strategy
We developed the strategy for identifying candidates for a 12-hour period of observation to exclude acute myocardial infarction from an analysis of data on 976 patients who were enrolled in the Chest Pain Study from December 1983 to October 15, 1984. These patients were a subgroup of those whose data had previously been analyzed to demonstrate the safety of a 24-hour period of observation.14 In developing the new strategy for a shorter observation period, we considered patients to be candidates for a 12-hour observation if they were classified as being at low (≤7 percent) risk of acute myocardial infarction according to a previously described multivariate algorithm (Fig. 1Figure 1
Prospectively Validated Multivariate Algorithm for the Prediction of a Patient's Risk of Acute Myocardial Infarction on the Basis of Emergency Room Data.). The algorithm was derived from an analysis of clinical data from 1379 patients who were evaluated at the emergency rooms of Yale–New Haven Hospital in 1977 or Brigham and Women's Hospital in 1978 and 1980–1982. Recursive partitioning analysis was used to divide patients into subgroups with different risks of acute myocardial infarction on the basis of their emergency room clinical data. Details of the identification of these patients and the derivation of this protocol have been published previously.2 None of the patients whose data were used to develop the algorithm are included in the population described in this report.Data from the entire 976-patient derivation set were used to determine the sensitivity and specificity for detecting acute myocardial infarction of the 12-hour and 24-hour periods of observation by determining the interval (in minutes) between a patient's arrival in the emergency room and the time at which the patient met the criteria for acute myocardial infarction. For patients with abnormal levels of cardiac enzymes, the time at which the blood sample was drawn was considered the time at which the patient met the diagnostic criteria.
The sensitivity of an observation period for detecting infarction was defined as the number of patients with acute myocardial infarction who had either abnormal levels of cardiac enzymes or recurrent ischemic pain during the period divided by the total number of patients with acute myocardial infarction. Specificity was defined as the number of patients without acute myocardial infarction who had neither abnormal enzyme levels nor recurrent ischemic pain during the period divided by the total number of patients without acute myocardial infarction.
To develop a Bayesian predictive model, we used the sensitivity and specificity of each observation period for detecting infarction to calculate the likelihood ratio associated with not having an abnormal enzyme level or recurrent ischemic pain during the period. The likelihood ratio was defined as the percentage of patients with acute myocardial infarction who did not have abnormal enzyme levels or recurrent ischemic pain during the observation period divided by the percentage of patients without infarction who did not have abnormal enzyme levels or recurrent ischemic pain. The predictive model was based on the odds-ratio expression of Bayes' theorem24:
posterior odds = prior odds × likelihood ratio.
To determine the probability of myocardial infarction for a patient who had neither abnormal enzyme levels after admission nor recurrent ischemic pain, one would thus calculate the patient's prior (i.e., preadmission) odds of infarction using the patient's emergency room clinical data and the multivariate algorithm (Fig. 1). Multiplying these odds by the likelihood ratio for a "negative" 12-hour period of observation would allow calculation of the patient's posterior odds (i.e., odds after 12 hours of observation) and probability of infarction (Fig. 2).
Prospective Validation
We tested the ability of this model to identify candidates for a 12-hour period of observation in an independent population of 2684 patients whose data had not been used in either the development of the emergency room algorithm (Fig. 1) or the initial evaluation of the diagnostic performance of the 12- and 24-hour periods of observation. Patients were considered to be candidates for the 12-hour period if they were classified as being at low risk of acute myocardial infarction on the basis of the emergency room algorithm. In these patients, the prior probability of infarction was considered to be the prevalence of infarction among the low-risk patients in the derivation set.
We evaluated the consequences of a 12-hour period of observation in the patients who were considered to be at low risk of infarction by determining the rates of acute myocardial infarction, major complications, and selected major procedures in patients who did not meet the criteria for infarction or have recurrent ischemic chest pain during that period. For patients who were considered to be at higher risk of infarction on the basis of their emergency room data, we evaluated the accuracy of the 12-hour period of observation and a previously described 24-hour strategy.12 Most of the patients who were deemed to be candidates for a 12-hour observation period were not actually transferred from the coronary care unit (or other monitored setting) after 12 hours; the length of each patient's stay in the coronary care unit was determined by that patient's physician.
We compared the clinical characteristics and triage decisions during hospitalization of the high- and low-risk patients in both the derivation and validation sets using the chi-square test for categorical variables and Student's t-test for age.
Results
Strategy Developed with Data on Patients in the Derivation Set
Acute Myocardial Infarctions in Low-Risk and High-Risk Patients
Of the 976 patients who were admitted to the hospital before October 15, 1984, 883 (90 percent) were admitted to the coronary care unit, 316 (32 percent) were found to have acute myocardial infarction, and 50 (5 percent) died during hospitalization (Table 1). After using the multivariate algorithm in Figure 1 to analyze the clinical data from the emergency room evaluations of these 976 patients, we classified 307 (31 percent) as being at low (≤7 percent) risk of acute myocardial infarction. Although the overall rate of myocardial infarction in patients whom we identified as being at low risk on the basis of this algorithm was about 2 percent,2 in the group of low-risk patients who were admitted, 23 (7 percent) were subsequently found to have acute myocardial infarction, and 71 (23 percent) had unstable angina (Table 2). Of the 669 patients in the derivation set who were classified as being at high (>7 percent) risk of infarction, 293 (44 percent) actually had acute myocardial infarction and 139 (21 percent) had unstable angina.
Data on the patients in the derivation set thus indicated that the probability of acute myocardial infarction among patients who were admitted to these seven hospitals for the evaluation of acute chest pain and who had serial enzyme testing was 7 percent if the patients' emergency room data defined them as being at low risk for the diagnosis and 44 percent if their emergency room data defined them as at high risk. We used these rates as prior probabilities for our predictions regarding the low-risk and high-risk patients in the validation set.
Sensitivity and Specificity of the Periods of Observation
Of the 316 patients in the derivation set who had acute myocardial infarction, 304 (96 percent) met the clinical criteria for this diagnosis or had recurrent ischemic pain during the first 12 hours of hospitalization, and 314 (99 percent) met the clinical criteria or had recurrent pain during the first 24 hours (Table 3). A strategy of keeping patients in monitored settings for 12 hours, with further monitored observation for those with clinical evidence of acute myocardial infarction or recurrent ischemic chest pain, would thus have allowed the transfer or discharge of 4 percent of the patients who were ultimately found to have infarction. Observation for 24 hours would have allowed the transfer or discharge of only 1 percent of the patients with infarction.
These strategies would also have led to the continued observation of patients without infarction who had recurrent ischemic chest pain or abnormalities in cardiac enzyme levels not judged by blinded reviewers to represent myocardial necrosis. Of the 660 patients without infarction in the derivation set, 517 (78 percent) had neither enzyme abnormalities nor recurrent ischemic chest pain during the first 12 hours, and 492 (75 percent) had neither during the first 24 hours (Table 3). Thus, the proposed 12-hour strategy would have led to the continued observation in a monitored setting of 22 percent of the patients without infarction. A 24-hour strategy would have led to the continued observation of 25 percent of the patients without infarction.
Predicted Probabilities of Acute Myocardial Infarction after 12 and 24 Hours of Observation
Using the sensitivity and specificity of 12-hour and 24-hour periods of observation, we were able to calculate the likelihood ratios for acute myocardial infarction associated with the absence of both enzyme abnormalities and recurrent ischemic pain. For both periods, the absence of abnormalities greatly reduced the risk of acute myocardial infarction: for a 12-hour period of observation without enzyme abnormalities or recurrent pain the likelihood ratio was 0.05, and for a similar 24-hour period the likelihood ratio was 0.008 (Table 3).
We used these likelihood ratios in a Bayesian model to predict the probability of acute myocardial infarction in patients with various initial (i.e., emergency room) probabilities of infarction who did not have abnormal enzyme levels or recurrent ischemic pain during the first 12 or 24 hours (Fig. 2). Thus, patients with a low prior risk of acute myocardial infarction according to their emergency room data would be predicted to have only a 0.4 percent probability of infarction after 12 hours of observation if they did not have enzyme abnormalities or recurrent ischemic chest pain. By comparison, patients at higher risk, who had a 44 percent prior probability of infarction, would still be predicted to have a 4 percent probability of infarction. However, if the patients at higher risk were observed for a full 24 hours and did not have enzyme abnormalities or recurrent chest pain, their predicted probability of infarction would decrease to 1 percent, as indicated by the lower curve (Fig. 2).
Prospective Validation of the 12-Hour Strategy
The 2684 patients in the validation set included 806 (30 percent) who had acute myocardial infarction and 119 (4 percent) who died during hospitalization (Table 1). Using the multivariate predictive algorithm (Fig. 1), we classified 957 (36 percent) as being at low risk of acute myocardial infarction on the basis of their emergency room data; they would thus have been candidates for a 12-hour period of observation. Of these candidates, 693 (72 percent) were initially assigned by triage to the coronary care unit. The rates of acute myocardial infarction (7 percent) and death during hospitalization ( 1 percent) among the low-risk patients in the validation set were similar to those among the low-risk patients in the derivation set (Table 2).
Of the 957 low-risk patients who were candidates for a 12-hour observation period, 771 (81 percent) did not have abnormal enzyme levels or recurrent ischemic chest pain during that period, and would thus have been considered for transfer out of a monitored setting, transfer to a lower level of care, or further evaluation with exercise testing. Of these 771 patients, 509 (66 percent) were actually admitted to coronary care units; these patients had a mean length of stay in the coronary care unit of 1.8±2.7 days. The 771 patients included 4 with acute myocardial infarction (nonfatal in all 4), for a rate of 0.5 percent, as compared with the 0.4 percent rate predicted by the Bayesian model (Table 3 and Fig. 2). In contrast, 36 (5 percent) of the 797 patients at higher risk who had not met the criteria for infarction within 12 hours were ultimately found to have this diagnosis (P<0.0001), a rate very similar to the predicted rate of 4 percent.
Rates of major complications and procedures were low in the 771 patients who were candidates for the 12-hour period of observation and who did not have enzymatic evidence of myocardial infarction or recurrent ischemic pain during that period (Fig. 3). There were no deaths during the first two days after admission, but 5 patients (0.6 percent) died from complications of ischemic heart disease three days or more after admission. Three of the patients (0.4 percent), all with congestive heart failure, had cardiac arrest three to five days after admission. One arrest occurred 10 hours after coronary angiography for unstable angina. A second occurred in a patient in whom atrial tachyarrhythmias developed on the fourth hospital day, making readmission to an intensive care unit necessary, and complete heart block developed on the fifth day. The third arrest occurred on the fourth hospital day in a patient who had been admitted with digitalis toxicity and whose condition had remained unmonitored since day 2. The other two deaths from cardiac causes were in a patient who died during Coronary Artery bypass surgery one week after admission and in a patient who was hospitalized for a prolonged period because of multiple medical problems and who had a fatal acute myocardial infarction six months after admission. All five deaths from cardiac causes occurred after the patients had been transferred out of the coronary or intensive care unit. Three other patients died from causes other than ischemic heart disease: one with a pulmonary embolism (five days after admission), one with a cerebrovascular accident (six days after admission), and one with a ruptured abdominal aortic aneurysm (nine days after admission).
Among the 763 candidates for a 12-hour observation period who survived hospitalization, nonfatal life-threatening complications occurred in only 3 (0.4 percent) (Fig. 3). In one, cardiogenic shock due to pericardial tamponade developed during hemodialysis for uremic pericarditis on the first day of hospitalization; another had a ventricular-fibrillation arrest after surgery to replace aortic and mitral valves 12 days after admission; and the third, who had severe congestive heart failure, had a ventricular-fibrillation arrest 10 days after admission. Two patients had nonfatal pulmonary emboli, and there were no cases of aortic dissections. Only eight patients underwent cardiac catheterization during the first three days after admission followed by coronary revascularization (coronary-artery bypass operations in five and percutaneous transluminal coronary angioplasty in three) during the same hospitalization. A review of the cases of the patients who had major complications or underwent urgent cardiac catheterization did not identify any for whom continued observation in the coronary care unit during the period between 12 and 24 hours after admission would have been likely to improve outcomes.
Evaluation of a 24-Hour Strategy for Patients at Higher Risk
Of the 1727 patients in the validation set whose risk of acute myocardial infarction was too high for them to be considered candidates for the 12-hour period of observation, 739 (43 percent) had acute myocardial infarction. Our previously described strategy14 of monitoring patients for 24 hours, with longer periods of monitoring for patients with recurrent ischemic chest pain, would have led to the detection of 728 (99 percent) of the 739 infarctions among patients at higher risk. The 1.5 percent rate of acute myocardial infarction (n = 11) among the 741 patients at higher risk who did not have abnormal enzyme levels or recurrent ischemic chest pain during the first 24 hours was similar to the rate of 1 percent predicted by the Bayesian model.
Discussion
Although previous investigations have suggested that 24 hours is the minimal safe period of observation in a monitored setting for patients with suspected acute myocardial infarction,14 , 15 this report demonstrates that a large group of patients who are admitted for the evaluation of acute pain can be identified as appropriate candidates for a 12-hour rule-out period on the basis of their emergency room clinical data. In this population of 3660 patients from community and university hospitals, all of whom were admitted and underwent serial testing of cardiac enzyme levels, more than one third of the patients (1264) would have been considered appropriate subjects for a 12-hour period of observation. In this study, most of the patients were not actually transferred from the monitored setting after 12 hours; thus, our data do not speak directly to the consequences of such a transfer.
Within the subgroup of these candidates who did not have enzymatic evidence of infarction or recurrent ischemic chest pain during the first 12 hours after admission, the 0.5 percent rate of acute myocardial infarction was slightly lower than the 1.4 percent rate of infarction found in patients who are discharged from the emergency room.3 Only 5 of these 771 patients (0.6 percent) died of cardiac causes during hospitalization, and there were no deaths during the first three days. Rates of major complications and procedures were low in this population, and the timing of these complications did not indicate that continued observation in the coronary care unit during the period between 12 and 24 hours after admission would have allowed physicians to anticipate them or change their outcomes.
Our analysis emphasizes the use of guidelines for rule-out protocols for myocardial infarction; it does not indicate that patients at lower risk should be discharged from the hospital after 12 hours if they do not have clinical evidence of acute myocardial infarction. Some patients have unstable angina and may require continued titration of medications, supervised ambulation, or cardiac catheterization. A low risk of acute myocardial infarction does not necessarily imply a low risk of unstable angina; 28 percent of the patients in both the low-risk and high-risk groups had unstable angina. Other patients may have problems other than ischemic heart disease, such as cardiac arrhythmias, acute gastrointestinal bleeding, or respiratory failure, that require continued intensive care.
Two recent studies have demonstrated, however, that identifying low-risk patients who may be appropriate for transfer out of the coronary care unit after 24 hours leads to significant decreases in lengths of stay in the coronary care unit.12 , 13 Identifying patients for whom a 12-hour period of observation is safe and effective can be expected to lead to further decreases in lengths of stay in the coronary care unit and more expeditious evaluations for many patients, because decisions to transfer or discharge a patient or to perform an exercise test are usually made during the morning. Patients who present with acute chest pain during the afternoon or at night therefore frequently remain in the coronary care unit for a second night after their 24-hour observation period is over.
The previous investigations that have described management strategies for patients 24 hours after admission included smaller numbers of patients and used only clinical data from the postadmission course, such as cardiac enzyme levels, electrocardiograms, and the occurrence of complications.14 , 15 In this prospective validation of a strategy to identify candidates for a shorter period of observation, low-risk patients from community and teaching hospitals who were appropriate for transfer after 12 hours were identified by integrating postadmission data with clinical data from the emergency room.
In our Bayesian approach, the emergency room clinical data are used to estimate a pretest probability of acute myocardial infarction, and the postadmission data from the first 12 hours are analogous to a test result that is used to modify that pretest probability. Negative test results after 24 hours have more meaning than negative results after 12 hours; a 24-hour period of observation can be considered a better (and more costly) test than a 12-hour period. However, as demonstrated by this analysis, the 12-hour period is sufficient to exclude myocardial infarction with reliability in patients with a low pretest probability of acute myocardial infarction. Other approaches besides the algorithm used in this investigation could be used to classify patients as at low risk and therefore eligible for the strategy of 12-hour observation, provided that these approaches are known to yield valid estimates of a patient's risk status.
By extension, it would be reasonable to expect that a period of observation longer than 24 hours might be necessary to exclude infarction in patients with a higher risk of this diagnosis at the time of the emergency room evaluation. However, previously described strategies14 , 15 were able to identify patients with a low rate of subsequent infarction among those at high risk in this study population after 24 hours, indicating that this period is generally sufficient for patients who are not candidates for the shorter observation period.
Since patients with suspected acute myocardial infarction who are discharged from the hospital without infarction have long-term prognoses that are similar to those of patients with infarction,25 it could also be argued that low-risk patients who do not have evidence of infarction after 12 hours should normally have further evaluation, such as formal or informal exercise testing, before discharge from the hospital. Our protocol was designed to identify only patients with myocardial infarction, not other acute coronary syndromes, such as unstable angina. Our data provide guidelines for discontinuing rule-out protocols for myocardial infarction, so that these other issues in evaluation and treatment can be addressed in an efficient and timely manner.
Supported in part by grants from the Robert Wood Johnson Foundation (12543), the John A. Hartford Foundation (83102–2H), and the Agency for Health Care Policy and Research (HS 06452). Dr. Lee is the recipient of an Established Investigator Award (900119) from the American Heart Association, and Dr. Rouan is a Teaching and Research Scholar of the American College of Physicians and was supported in part by a grant to the Training Program in Clinical Effectiveness from the W.K. Kellogg Foundation.
Source Information
From Brigham and Women's Hospital (T.H.L., E.F.C., M.C.W., L.G.), Harvard Medical School (T.H.L., G.J., L.G.), and the Harvard School of Public Health (T.H.L., E.F.C.), all in Boston; Yale–New Haven Hospital and Yale University School of Medicine, New Haven, Conn. (D. A.B.); and the University of Cincinnati Hospital and the University of Cincinnati, Cincinnati (G.W.R.). Address reprint requests to Dr. Lee at the Division of Clinical Epidemiology, Department of Medicine, Brigham and Women's Hospital, 75 Francis St., Boston, MA 02115.
Appendix: The Chest Pain Study Group
The Chest Pain Study Group includes Lee Goldman, M.D. (co-principal investigator), Thomas H. Lee, M.D., E. Francis Cook, Sc.D., Monica C. Weisberg, R.N., Karen Daley, R.N., and Barbara C. Rosen, B.A., Brigham and Women's Hospital, Boston; George Terranova, M.D. (site director), Carol Stasiulewicz, P.A., and David Copen, M.D., Danbury Hospital, Danbury, Conn.; Alan Brandt, M.D. (site director), and Jay Walshon, M.D., Milford Hospital, Milford, Conn.; Louis Gottlieb, M.D. (site director), St. Mary's Hospital, Waterbury, Conn.; Gregory W. Rouan, M.D. (site director), Jerris R. Hedges, M.D., Robert Toltzis, M.D., and Beth Goldstein-Wayne, R.N., University of Cincinnati Hospital, Cincinnati; Michael Kobernick, M.D. (site director), Daniel Jones, B.A., and Carolyn Guidot, M.D., William Beaumont Hospital, Royal Oak, Mich.; and Donald A. Brand, Ph.D. (co-principal investigator), Denise Acampora, M.P.H., John Mellors, M.D., Kathryn Trainor, M.S., Rita M. Jakubowski, R.N., and Susan Healy, R.N., Yale–New Haven Hospital, New Haven, Conn.
- References
References
1
Lee
Web of Science | Medline2
Goldman
Full Text | Web of Science | Medline3
Lee
CrossRef | Web of Science | Medline4
Lee
CrossRef | Web of Science | Medline5
Lee
CrossRef | Web of Science | Medline6
Tierney
CrossRef | Web of Science | Medline7
Pozen
Full Text | Web of Science | Medline8
Brush
Full Text | Web of Science | Medline9
Slater
CrossRef | Web of Science | Medline10
Rouan
CrossRef | Web of Science | Medline11
Weingarten
CrossRef | Web of Science | Medline12
Weingarten
Web of Science | Medline13
Eagle
CrossRef | Web of Science | Medline14
Lee
Web of Science | Medline15
Mulley
Full Text | Web of Science | Medline16
Lee
Web of Science | Medline17
Solomon
CrossRef | Web of Science | Medline18
Beamer
CrossRef | Web of Science | Medline19
Cunningham
CrossRef | Web of Science | Medline20
Lee
CrossRef | Web of Science | Medline21
White
CrossRef | Web of Science | Medline22
Lee
CrossRef | Web of Science | Medline23
Fiebach
CrossRef | Web of Science | Medline24
McNeil
Full Text | Web of Science | Medline25
Schroeder
Full Text | Web of Science | Medline
- Citing Articles (67)
Citing Articles
1
J. Douglas Kirk, Michael Kontos, Deborah B. Diercks. (2011) Upstream treatment of acute coronary syndrome in the ED. The American Journal of Emergency Medicine 29:4, 446-456
CrossRef2
Leo Bossaert, Robert E. O’Connor, Hans-Richard Arntz, Steven C. Brooks, Deborah Diercks, Gilson Feitosa-Filho, Jerry P. Nolan, Terry L. Vanden Hoek, Darren L. Walters, Aaron Wong, Michelle Welsford, Karen Woolfrey. (2010) Part 9: Acute coronary syndromes. Resuscitation 81:1, e175-e212
CrossRef3
Vikas Singh, Pedro Martinezclark, Mario Pascual, Eric Scot Shaw, William W. OʼNeill. (2010) Cardiac biomarkers – the old and the new: a review. Coronary Artery Disease 21:4, 244-256
CrossRef4
W. Frank Peacock, Karina M. Soto-Ruiz. (2009) Risk stratification for suspected acute coronary syndromes and heart failure in the emergency department. Acute Cardiac Care 11:3, 138-145
CrossRef5
Judd E Hollander. 2008. Optimal Strategies for Prehospital and Emergency Department Evaluation of Patients with STEMI. , 265-279.
CrossRef6
Aaron M. Brown, Keara L. Sease, Jennifer L. Robey, Frances S. Shofer, Judd E. Hollander. (2007) The Impact of B-Type Natriuretic Peptide in Addition to Troponin I, Creatine Kinase-MB, and Myoglobin on the Risk Stratification of Emergency Department Chest Pain Patients With Potential Acute Coronary Syndrome. Annals of Emergency Medicine 49:2, 153-163
CrossRef7
Michael J. Gallagher, Michael A. Ross, Gilbert L. Raff, James A. Goldstein, William W. O’Neill, Brian O’Neil. (2007) The Diagnostic Accuracy of 64-Slice Computed Tomography Coronary Angiography Compared With Stress Nuclear Imaging in Emergency Department Low-Risk Chest Pain Patients. Annals of Emergency Medicine 49:2, 125-136
CrossRef8
Jin H. Han, Christopher J. Lindsell, Alan B. Storrow, Samuel Luber, James W. Hoekstra, Judd E. Hollander, W. Franklin Peacock, Charles V. Pollack, W. Brian Gibler. (2007) The Role of Cardiac Risk Factor Burden in Diagnosing Acute Coronary Syndromes in the Emergency Department Setting. Annals of Emergency Medicine 49:2, 145-152.e1
CrossRef9
Neal Benowitz. 2006. Medical Complications of Drug Abuse. , 597-694.
CrossRef10
Oleg Gorelik, Dorit Almoznino-Sarafian, Israel Yarovoi, Irena Alon, Miriam Shteinshnaider, Shulamit Chachashvily, David Modai, Natan Cohen. (2006) Patient-related variables predicting acute coronary syndrome following admission for chest pain of possible coronary origin. Coronary Artery Disease 17:1, 15-21
CrossRef11
Chih-Hao Lin, Shey-Ying Chen, Wei-Tien Chang, Shyr-Chyr Chen, Wen-Jone Chen. (2005) Cardiac rupture after minor blunt chest injury as a result of occult recent myocardial infarction. Resuscitation 67:1, 149-150
CrossRef12
, Judd E. Hollander, Andra L. Blomkalns, Gerard X. Brogan, Deborah B. Diercks, John M. Field, J. Lee Garvey, W. Brian Gibler, Timothy D. Henry, James W. Hoekstra, Brian R. Holroyd, Yuling Hong, J. Douglas Kirk, Brian J. O'Neil, Raymond E. Jackson. (2004) Standardized Reporting Guidelines for Studies Evaluating Risk Stratification of ED Patients with Potential Acute Coronary Syndromes. Academic Emergency Medicine 11:12, 1331-1340
CrossRef13
Deborah B. Diercks, Judd E. Hollander, Frank Sites, J. Douglas Kirk. (2004) Derivation and Validation of a Risk Stratification Model to Identify Coronary Artery Disease in Women Who Present to the Emergency Department with Potential Acute Coronary Syndromes. Academic Emergency Medicine 11:6, 630-634
CrossRef14
Roberto Bassan, Lucia Pimenta, Marcelo Scofano, José Francisco Soares. (2004) Accuracy of a Neural Diagnostic Tree for the Identification of Acute Coronary Syndrome in Patients With Chest Pain and No ST-Segment Elevation. Critical Pathways in Cardiology: A Journal of Evidence-Based Medicine 3:2, 72-78
CrossRef15
Deborah B. Diercks, Judd E. Hollander, Frank Sites, J. Douglas Kirk. (2004) Derivation and Validation of a Risk Stratification Model to Identify Coronary Artery Disease in Women Who Present to the Emergency Department with Potential Acute Coronary Syndromes. Academic Emergency Medicine 11:5, 630-634
CrossRef16
Roberto Bassan, Lucia Pimenta, Marcelo Scofano, Roberto Gamarski, André Volschan. (2004) Probability Stratification and Systematic Diagnostic Approach for Chest Pain Patients in the Emergency Department. Critical Pathways in Cardiology: A Journal of Evidence-Based Medicine 3:1, 1-7
CrossRef17
Sharon E. Mace. (2004) Continuous Quality Improvement for the Clinical Decision Unit. Journal for Healthcare Quality 26:1, 29-36
CrossRef18
A Bellou, J.-D de Korwin, J Bouget, F Carpentier, V Ledoray, J Kopferschmitt, H Lambert. (2003) Place des services d’urgences dans la régulation des hospitalisations publiques. La Revue de Médecine Interne 24:9, 602-612
CrossRef19
Graham S. Hillis, Pamela Taggart, Delana Wardlaw, Lorraine Hillis, Ning Zhao, William C. Dalsey, Antoinette Mangione. (2003) The relative utility of cardiac troponin i, creatine kinase-mb mass , and myosin light chain-1 in the long-term risk stratification of patients with chest pain. Clinical Cardiology 26:3, 147-152
CrossRef20
Weber, Jim Edward, Shofer, Frances S., Larkin, G. Luke, Kalaria, Amit S., Hollander, Judd E., . (2003) Validation of a Brief Observation Period for Patients with Cocaine-Associated Chest Pain. New England Journal of Medicine 348:6, 510-517
Full Text21
Alex F. Manini, Michael A. Gisondi, Theresa M. Vlugt, Donald H. Schreiber. (2002) Adverse Cardiac Events in Emergency Department Patients with Chest Pain Six Months after a Negative Inpatient Evaluation for Acute Coronary Syndrome. Academic Emergency Medicine 9:9, 896-902
CrossRef22
Francesco Pelliccia, Cinzia Cianfrocca, Giuseppe Marazzi, Mario Pagliei, Marcello Mariani, Giuseppe M. C. Rosano. (2002) Continuous 12-lead st-segment monitoring improves identification of low-risk patients with chest pain and a worse in-hospital outcome. Clinical Cardiology 25:2, 57-62
CrossRef23
PHILIP V. SCRIBANO, JAMES F. WILEY, KENNETH PLATT. (2001) Use of an observation unit by a pediatric emergency department for common pediatric illnesses. Pediatric Emergency Care 17:5, 321-323
CrossRef24
Alex Limkakeng, W. Brian Gibler, Charles Pollack, James W. Hoekstra, Frank Sites, Frances S. Shofer, Brian Tiffany, Eric Wilke, Judd E. Hollander. (2001) Combination of Goldman Risk and Initial Cardiac Troponin I for Emergency Department Chest Pain Patient Risk Stratification. Academic Emergency Medicine 8:7, 696-702
CrossRef25
Brian J. O'Neil. (2001) Another Attempt at Easing Chest Pain: Ours and Theirs. Academic Emergency Medicine 8:7, 752-753
CrossRef26
Harald Herkner, Ulla Waldenhofer, Anton N. Laggner, Marcus Müllner, Elisabeth Oschatz, Susanne Spitzauer, Gunnar Gamper, Andreas Bur, Michael M. Hirschl. (2001) Clinical application of rapid quantitative determination of cardiac troponin-T in an emergency department setting. Resuscitation 49:3, 259-264
CrossRef27
Gary B. Green, Erik Dehlinger, Thai S. McGrievey, Dai J. Li, Kerrie A. Jones, Gabor D. Kelen, Daniel W. Chan. (2000) CK-MB isoforms for early risk stratification of emergency department patients. Clinica Chimica Acta 300:1-2, 57-73
CrossRef28
Eugene Braunwald, Elliott M Antman, John W Beasley, Robert M Califf, Melvin D Cheitlin, Judith S Hochman, Robert H Jones, Dean Kereiakes, Joel Kupersmith, Thomas N Levin, Carl J Pepine, John W Schaeffer, Earl E Smith, David E Steward, Pierre Theroux, Raymond J Gibbons, Joseph S Alpert, Kim A Eagle, David P Faxon, Valentin Fuster, Timothy J Gardner, Gabriel Gregoratos, Richard O Russell, Sidney C Smith. (2000) ACC/AHA guidelines for the management of patients with unstable angina and non–st-segment elevation myocardial infarction. Journal of the American College of Cardiology 36:3, 970-1062
CrossRef29
Samantha MaWhinney, Elizabeth R Brown, Janet Malcolm, Catherine VillaNueva, Bertron M Groves, Robert A Quaife, JoAnn Lindenfeld, Bradley A Warner, Karl E Hammermeister, Frederick L Grover, A.Laurie W Shroyer. (2000) Identification of risk factors for increased cost, charges, and length of stay for cardiac patients. The Annals of Thoracic Surgery 70:3, 702-710
CrossRef30
Gary B. Green, George W. Skarbek-Borowski, Daniel W. Chan, Gabor D. Kelen. (2000) Myoglobin for Early Risk Stratification of Emergency Department Patients with Possible Myocardial Ischemia. Academic Emergency Medicine 7:6, 625-636
CrossRef31
Gordon A Ewy, Joseph P Ornato. (2000) Emergency cardiac care: introduction. Journal of the American College of Cardiology 35:4, 825-880
CrossRef32
Robbert J. de Winter, Jeroen G. Lijmer, Rudolph W. Koster, Frans J. Hoek, Gerard T. Sanders. (2000) Diagnostic accuracy of myoglobin concentration for the early diagnosis of acute myocardial infarction. Annals of Emergency Medicine 35:2, 113-120
CrossRef33
Gunnar Gustafsson, Mikael Dellborga. (2000) Early Diagnosis and Exclusion of Acute Myocardial Infarction by Two Hours' Vector-ECG and Determination of Either Myoglobin or CK-mb. Scandinavian Cardiovascular Journal 34:2, 172-177
CrossRef34
Judd E Hollander, M.Ranu Muttreja, Margaret R Dalesandro, Frances S Shofer. (1999) Risk stratification of emergency department patients with acute coronary syndromes using P-Selectin. Journal of the American College of Cardiology 34:1, 95-105
CrossRef35
William R Lewis, Ezra A Amsterdam, Samuel Turnipseed, J.Douglas Kirk. (1999) Immediate exercise testing of low risk patients with known coronary artery disease presenting to the emergency department with chest pain. Journal of the American College of Cardiology 33:7, 1843-1847
CrossRef36
JAMES A. TRIPPI, KAMTHORN S. LEE. (1999) Dobutamine Stress Tele-Echocardiography as a Clinical Service in the Emergency Department to Evaluate Patients With Chest Pain. Echocardiography 16:2, 179-185
CrossRef37
Newby, L. Kristin, Mark, Daniel B., . (1998) The Chest-Pain Unit — Ready for Prime Time?. New England Journal of Medicine 339:26, 1930-1932
Full Text38
James H. Stein, Alex Neumann, Lynn M. Preston, Betsy J. Vandenberg, Joseph E. Parrillo, James E. Calvin, Richard H. Marcus. (1998) Improved risk stratification in unstable angina: Identification of patients at low risk for in-hospital cardiac events by admission echocardiography. Clinical Cardiology 21:10, 725-730
CrossRef39
Sherif Iskander, Ami E Iskandrian. (1998) Risk assessment using single-photon emission computed tomographic technetium-99m sestamibi imaging. Journal of the American College of Cardiology 32:1, 57-62
CrossRef40
Robert Roberts. (1998) Early diagnosis of myocardial infarction with MB CK isoforms. Clinica Chimica Acta 272:1, 33-45
CrossRef41
Hamm, Christian W., Goldmann, Britta U., Heeschen, Christopher, Kreymann, Georg, Berger, Jürgen, Meinertz, Thomas, . (1997) Emergency Room Triage of Patients with Acute Chest Pain by Means of Rapid Testing for Cardiac Troponin T or Troponin I. New England Journal of Medicine 337:23, 1648-1653
Full Text42
James A Trippi, Kamthorn S Lee, Greg Kopp, David R Nelson, King G Yee, William H Cordell. (1997) Dobutamine Stress Tele-Echocardiography for Evaluation of Emergency Department Patients With Chest Pain. Journal of the American College of Cardiology 30:3, 627-632
CrossRef43
R Kennedy. (1997) An artificial neural network system for diagnosis of acute myocardial infarction (AMI) in the accident and emergency department: evaluation and comparison with serum myoglobin measurements. Computer Methods and Programs in Biomedicine 52:2, 93-103
CrossRef44
Miguel A. Gomez, Jeffrey L. Anderson, Labros A. Karagounis, Joseph B. Muhlestein, F. Bruce Mooders. (1996) An emergency department-based protocol for rapidly ruling out myocardial ischemia reduces hospital time and expense: Results of a randomized study (ROMIO). Journal of the American College of Cardiology 28:1, 25-33
CrossRef45
Goldman, Lee, Cook, E. Francis, Johnson, Paula A., Brand, Donald A., Rouan, Gregory W., Lee, Thomas H., . (1996) Prediction of the Need for Intensive Care in Patients Who Come to Emergency Departments with Acute Chest Pain. New England Journal of Medicine 334:23, 1498-1504
Full Text46
Sgarbossa, Elena B., Pinski, Sergio L., Barbagelata, Alejandro, Underwood, Donald A., Gates, Kathy B.Topol, Eric J., Califf, Robert M., Wagner, Galen S., . (1996) Electrocardiographic Diagnosis of Evolving Acute Myocardial Infarction in the Presence of Left Bundle-Branch Block. New England Journal of Medicine 334:8, 481-487
Full Text47
Bertil Lindahl. (1996) Biochemical Markers of Myocardial Damage for Early Diagnosis and Prognosis in Patients with Acute Coronary Syndromes. Upsala Journal of Medical Sciences 101:3, 193-232
CrossRef48
C Estrada. (1995) Role of telemetry monitoring in the non-intensive care unit. The American Journal of Cardiology 76:12, 960-965
CrossRef49
Steven D. Pearson, Lee Goldman, E. John Orav, Edward Guadagnoli, Tomas B. Garcia, Paula A. Johnson, Thomas H. Lee. (1995) Triage decisions for emergency department patients with chest pain. Journal of General Internal Medicine 10:10, 557-564
CrossRef50
E.W.M. Grijseels, J.W. Deckers, A.W. Hoes, J.A.M. Hartman, E. van der Does, M.L. Simoons. (1995) Optimal use of coronary care units: A review. Progress in Cardiovascular Diseases 37:6, 415-421
CrossRef51
Raffaele Bugiardini, Alberico Borghi, Andrea Pozzati, Augusto Ruggeri, Paolo Puddu, Attilio Maseri. (1995) Relation of severity of symptoms to transient myocardial ischemia and prognosis in unstable angina. Journal of the American College of Cardiology 25:3, 597-604
CrossRef52
V. Bhayana, A.Ralph Henderson. (1995) Biochemical markers of myocardial damage. Clinical Biochemistry 28:1, 1-29
CrossRef53
Marshall H. Chin, E. Francis Cook, Thomas H. Lee, Lee Goldman. (1994) Correlates of major complications and mortality in patients presenting to the emergency department with chest pain and more than bibasilar rales. Journal of General Internal Medicine 9:12, 659-665
CrossRef54
Jean-Michel Gaspoz, Thomas H. Lee, Milton C. Weinstein, E.Francis Cook, Paula Goldman, Anthony L. Komaroff, Lee Goldman. (1994) Cost-effectiveness of a new short-stay unit to “rule out” acute myocardial infarction in low risk patients. Journal of the American College of Cardiology 24:5, 1249-1259
CrossRef55
Vipin Bhayana, Steve Cohoe, Thomas G. Pellar, George Jablonsky, A.Ralph Henderson. (1994) Combination (multiple) testing for myocardial infarction using myoglobin, creatine kinase-2 (mass), and troponin T. Clinical Biochemistry 27:5, 395-406
CrossRef56
Hamm, Christian W., . (1994) New Serum Markers for Acute Myocardial Infarction. New England Journal of Medicine 331:9, 607-608
Full Text57
Pauker, Stephen G.Kopelman, Richard I.. (1994) Invasive Interventions. New England Journal of Medicine 331:9, 601-605
Full Text58
Puleo, Peter R.Meyer, DeniseWathen, CherylTawa, Cyril B.Wheeler, SusanHamburg, Robert J.Ali, NadirObermueller, Steven D.Triana, Fernando J.Zimmerman, Janice L.Perryman, M. BenjaminRoberts, Robert. (1994) Use of a Rapid Assay of Subforms of Creatine Kinase MB to Diagnose or Rule Out Acute Myocardial Infarction. New England Journal of Medicine 331:9, 561-566
Full Text59
Judd E. Hollander, Robert S. Hoffman, Paul Gennis, Phillip Fairweather, Michael J. DiSano, David A. Schumb, James A. Feldman, Susan S. Fish, Sophia Dyer, Paul Wax, Chris Whelan, Evan Schwarzwald. (1994) Prospective Multicenter Evaluation of Cocaine-associated Chest Pain. Academic Emergency Medicine 1:4, 330-339
CrossRef60
Kirsten E. Fleischmann, Lee Goldman, Paul A. Robiolio, Richard T. Lee, Paula A. Johnson, E.Francis Cook, Thomas H. Lee. (1994) Echocardiographic correlates of survival in patients with chest pain. Journal of the American College of Cardiology 23:6, 1390-1396
CrossRef61
Steven D. Pearson, Lee Goldman, Tomas B. Garcia, E. Francis Cook, Thomas H. Lee. (1994) Physician response to a prediction rule for the triage of emergency department patients with chest pain. Journal of General Internal Medicine 9:5, 241-247
CrossRef62
James W. Hoekstra, Jerris R. Hedges, W. Brian Gibler, R. Mike Rubison, Ronald A. Christensen, . (1994) Emergency Department CK-MB: A Predictor of Ischemic Complications. Academic Emergency Medicine 1:1, 17-27
CrossRef63
Kevin W. Yinclinc, Lawson R. Wulsin, Lesley Mussio Arnold, Gregory W. Rouan. (1993) Estimated prevalences of panic disorder and depression among consecutive patients seen in an emergency department with acute chest pain. Journal of General Internal Medicine 8:5, 231-235
CrossRef64
Gary Naglie, Michel Silberfeld, Keith O'Rourke, Bruce Fried, Wendy Corber, Claire Bombardier, Allan Detsky. (1993) A randomized trial of a decisional aid for mental capacity assessments. Journal of Clinical Epidemiology 46:3, 221-230
CrossRef65
Ronald A. Schoenenberger, Martin Conzelmann, Ulrich C. Dubach, Juerg Schwander. (1992) Quality of emergency room triage of medical inpatients to an acute care clinic or chronic health care facilities. Journal of General Internal Medicine 7:3, 321-327
CrossRef66
(1991) Ruling Out Acute Myocardial Infarction. New England Journal of Medicine 325:17, 1249-1250
Full Text67
Eagle, Kim A., . (1991) Medical Decision Making in Patients with Chest Pain. New England Journal of Medicine 324:18, 1282-1283
Full Text
- Letters
