Advertisement

Feasibility and Reliability of Rapid Diagnosis of Myocardial Infarction

      ABSTRACT

      Background

      Prevailing hospital practice dictates a protracted phase of observation for patients with chest pain to establish or exclude the diagnosis of myocardial infarction. Early diagnosis of acute myocardial infarction may improve patient care and reduce both complications and hospital costs. A study was performed to investigate the feasibility of early diagnosis of myocardial infarction within the first 9 hours of the hospital stay.

      Methods

      The records of all patients admitted with chest pain within one calendar year were analyzed. The timing of creatine kinase-MB (CK-MB) quantification was determined with reference to the initial phlebotomy (time 0). An enzymatic diagnosis of myocardial infarction was assigned if any determination of CK-MB exceeded the upper limit of normal, and the diagnosis of each patient at or before 9 hours (early diagnosis) was compared to the ultimate diagnosis at 14 to 24 hours (final diagnosis) beyond initial assessment.

      Results

      Of the 528 included patients, 523 patients (99.1%) had identical early and final diagnostic outcomes; 5 patients (0.9%) had conflicting results. An early diagnosis of myocardial infarction was assigned to 195 of the 528 patients (36.9%). Of these, 190 achieved the diagnosis within 9 hours (sensitivity 97.4%). The negative predictive value was 98.5%.

      Conclusion

      Standard CK-MB mass measurements within 9 hours of arrival provided an accurate clinical assessment in>99% of the cases. The high sensitivity and negative predictive values suggest that early diagnosis of myocardial infarction is feasible and reliable.

      KEY INDEXING TERMS

      Each year, more than 5 million patients are evaluated for chest pain.
      • Nourjah P.
      National Hospital Ambulatory Medical Care Survey: 1997 emergency department summary.
      Hospital utilization is predominated by diagnostic testing for suspected, new myocardial infarction. Patients are admitted for an average stay of 1.9 days and $6 billion is spent annually in the United States to rule out myocardial infarction.
      • Barish R.A.
      • Doherty R.J.
      • Browne B.J.
      Reengineering the emergency evaluation of chest pain.
      When patients present to the hospital with the acute onset of chest pain, their evaluation incorporates a careful medical history, thorough physical examination, electrocardiographic assessment (ECG), and the performance of serial blood analyses. The patient’s history and physical exam may suggest the likelihood of cardiac damage but cannot be considered definitive. The ECG might confirm less than 50% of myocardial infarctions.
      • Gibler W.B.
      • Young G.P.
      • Hedges J.R.
      • et al.
      Acute myocardial infarction in chest pain patients with nondiagnostic ECGs: serial CK-MB sampling in the emergency department. The Emergency Medicine Cardiac Research Group.
      • Rude R.E.
      • Poole W.K.
      • Muller J.E.
      • et al.
      Electrocardiographic and clinical criteria for recognition of acute myocardial infarction based on analysis of 3,697 patients.
      • Behar S.
      • Schor S.
      • Kariv I.
      • et al.
      Evaluation of electrocardiogram in emergency room as a decision-making tool.
      Thus, in those patients who ultimately receive a diagnosis of myocardial infarction, and in the vast majority of patients for whom the diagnosis is ultimately excluded, serum analysis will be required to detect the diagnostic patterns of cardiac enzyme release associated with myocyte necrosis. In most hospitals, serum is collected serially over at least 12 to 16 hours. Clinical decisions (either discharge from the hospital with close follow-up or an immediate diagnostic study) are postponed until these laboratory values are available, resulting in prolonged and expensive hospitalization.
      Shortening the time required for evaluation of chest pain patients may have benefits beyond cost-savings and the earlier identification and treatment of myocardial infarction patients. Several studies have demonstrated that at least 4 to 8% of patients with early, evolving myocardial infarction are mistakenly discharged from the emergency department without the serial enzyme analysis that would enable an appropriate diagnosis.
      • McCarthy B.D.
      • Beshansky J.R.
      • D’Agostino R.B.
      • et al.
      Missed diagnoses of acute myocardial infarction in the emergency department: results from a multicenter study.
      • Schor S.
      • Behar S.
      • Modan B.
      • et al.
      Disposition of presumed coronary patients from an emergency room. A follow-up study.
      • Sigurdsson E.
      • Thorgeirsson G.
      • Sigvaldason H.
      • et al.
      Unrecognized myocardial infarction: epidemiology, clinical characteristics, and the prognostic role of angina pectoris. The Reykjavik Study.
      These patients have a significantly higher short-term mortality rate than those patients who are admitted and receive accurate diagnostic and therapeutic intervention.
      • Lee T.H.
      • Rouan G.W.
      • Weisberg M.C.
      • et al.
      Clinical characteristics and natural history of patients with acute myocardial infarction sent home from the emergency room.
      It can be conjectured that, if the time and expense required for serial blood testing were decreased, many patients who elude appropriate diagnostic evaluation might be more rigorously scrutinized.
      A modification of the current paradigm for chest pain evaluation, if it could safely and effectively shorten the in-hospital stay required to exclude myocardial infarction, would provide substantial cost savings, lead to earlier treatment of myocardial infarction and might reduce the number of patients whose diagnosis eludes detection. To investigate the feasibility of early laboratory diagnosis of myocardial infarction within the first 9 hours after admission, a retrospective study was performed.

      Methods

      Study Design

      All patients admitted to Stanford University Hospital in 1998 were reviewed. A total of 1698 patients were discharged with a billing diagnosis of chest pain, angina, ischemic heart disease, or myocardial infarction. Of the 1698 patients, 863 were admitted through the emergency department. These cases were analyzed in detail using electronic laboratory records and hospital charts. A total of 528 patients had sufficient creatine kinase-MB (CK-MB) determinations to establish both an early and a final enzymatic diagnosis of myocardial infarction (see below) and were thereby eligible for inclusion in this study. Troponin I and myoglobin measurements were also collected for secondary analysis. In addition, ECGs obtained during the first 24 hours of hospitalization for all patients with an enzymatic diagnosis of myocardial infarction were evaluated for signs of ischemia (T-wave inversion and ST-segment depression) or infarction (ST-segment depression, Q waves, and left bundle-branch block).

      Enzymatic Diagnosis of Myocardial Infarction

      The timing of CK-MB quantification was determined with reference to the initial phlebotomy (time 0). An enzymatic diagnosis of myocardial infarction was assigned if any determination of CK-MB exceeded the upper limit of normal for this assay. ‘Early’ diagnosis was arbitrarily defined as enzymatic diagnosis of myocardial infarction at or before 9 hours. ‘Final’ diagnosis was arbitrarily defined as an enzymatic diagnosis of myocardial infarction at any time up to 14 to 24 hours. To be included for analysis, patients must have had, at a minimum, determinations of 3 separate CK-MB values, with one at time 0, one within 9 hours, and a third at 14 to 24 hours.

      Cardiac Marker Assays

      CK-MB was measured using the OPUS-Magnum mass assay (Dade-Behring). A value of 5.0 μg/L was defined as the upper limit of normal. Troponin I and myoglobin were also measured using the OPUS-Magnum device with an upper limit of normal of 0.5 μg/L and 106 μg/L, respectively.

      Statistical Analysis

      The differences in age and proportion of men and women between the groups as well as the differences in age between men and women among patients with enzymatically diagnosed myocardial infarction were tested with the single factor ANOVA test. The sensitivity, specificity, and positive and negative predictive values of early enzymatic diagnosis were compared to the final diagnosis. Subgroup analysis was performed, excluding patients with ST-elevation myocardial infarction. The sensitivity and specificity of troponin I and myoglobin at each time point were also calculated with reference to the outcome for the final enzymatic diagnosis by CK-MB determination.

      Results

      The study population included only patients admitted through the emergency department with a diagnosis of chest pain, angina, ischemic heart, or myocardial infarction who had sufficient CK-MB data for analysis. Of the 528 included patients, 195 (36.9%) were assigned an enzymatic diagnosis of myocardial infarction. The patient demographics for this analysis are summarized in Table 1. Patients with an enzymatic diagnosis of myocardial infarction were much more likely to be male than those without an enzymatic diagnosis of myocardial infarction (70.3 versus 51.4%, p=0.00002). The average age of the patients with an enzymatic diagnosis of myocardial infarction was slightly higher than those without myocardial infarction (69.3 versus 66.8), but the difference was not statistically significant (p=0.06). Of note, the average age of women with enzymatic diagnosis of myocardial infarction was significantly higher than that of men (73.6 versus 67.5 years, p=0.004). ECGs in patients with an enzymatic diagnosis of myocardial infarction were normal in 10.3%. Q waves were present in 32.8%, ST-segment elevation in 41.5%, ST-segment depression in 23.1%, T-wave inversion in 36.9%, and left bundle-branch block in 2.6%.
      Table 1Clinical Characteristics of Patients
      All patients (n=528)MI (n=195)No MI (n=333)
      Age, mean±SD, yrs67.7±14.869.3±13.666.8±15.4
      Men, n (%)308 (58.3)137 (70.3)
      The difference is statistically significant (p=0.00002).
      171(51.4)
      The difference is statistically significant (p=0.00002).
      Women, n (%)220 (41.7)58(29.7)
      The difference is statistically significant (p=0.00002).
      162 (48.6)
      The difference is statistically significant (p=0.00002).
      MI, myocardial infarction.
      * The difference is statistically significant (p=0.00002).
      Of the 195 patients with a final enzymatic diagnosis of myocardial infarction, 109 had positive CK-MB values on admission (sensitivity 55.9%). Within 9 hours, 190 achieved the diagnostic criteria of at least one positive CK-MB value (sensitivity 97.4%). Only 5 patients had a delayed rise in CK-MB (enzymatic diagnosis late but not early). This yields a negative predictive value of 98.5%. The temporal sensitivity of enzymatic screening to diagnose acute myocardial infarction is displayed in Figure 1. As can be seen, after 8 hours, sensitivity exceeds 90%, and after 9 hours, sensitivity is over 97%. The specificity and positive predictive value are both 100% because, by our definition, the diagnosis of myocardial infarction was assigned to any patient with a diagnostic rise in CK-MB, either early or late. Statistical analysis was also performed after exclusion of the 81 patients with ST-elevation. Of the 114 remaining patients with an enzymatic diagnosis of myocardial infarction, 111 achieved diagnostic criteria within 9 hours. The sensitivity remained unchanged (97.4%) and the negative predictive value improved to 99.1% when excluding those patients with ST-elevation.
      Figure thumbnail gr1
      Figure 1Temporal sensitivity of enzymatic screening to diagnose acute myocardial infarction. The time is the number of hours from initial phlebotomy.
      The 5 patients with discordant values are described in Table 2. For 3 of the 5, a careful chart review disclosed an intervening in-hospital event that could explain the delayed enzyme rise. Patient 433 had an episode of recurrent chest pain after admission that might have accounted for later myocardial necrosis. Patient 199 had percutaneous coronary angioplasty and stent placement after admission. The intervention was complicated by documented distal embolization with complaints of chest pain and hypotension. The embolization was believed to have occasioned the late enzyme rise. Patient 502 also had angioplasty after admission but no specific events were documented. The maximum CK-MB for this patient was 5.2, marginally outside the normal range. Patient 152 had a single, very high CK-MB value in the setting of a normal total CK. This was believed to reflect a technical error. The one patient without any alternative explanation had a modest elevation in CK-MB to 6.33, but total CK at that time in the patient’s course was declining and, concurrently, the troponin I was normal.
      Table 2Five Patients with Enzymatic Diagnosis of MI Late but not Early: CK-MB Values and Clinical Information CK-MB values are in micrograms per liter
      Patient No.CK-MB (0 h)CK-MB (≤9 h)CK-MB (14–24 h)Clinical Information
      1521.030.923.9CK-MB of 23.9 believed to be an error as total CK was falling and was only 82.1 at the time.
      1991.841.5611.5The patient had angioplasty complicated by distal embolization, chest pain, and hypotension.
      4211.711.876.33Total CK was falling at the time of the elevated third CK-MB and a troponin I was normal.
      4332.374.79.79There was recurrent chest pain around the time of the second CK-MB measurement.
      5023.684.375.2The patient had an angioplasty between the first and second CK-MB measurements.
      A secondary analysis of the sensitivity and specificity of troponin I and myoglobin at each time point was also performed (Table 3). Serial troponin I measurements were available at each point for a relatively limited cohort of these patients. On admission, 513 of the 528 patients had troponin I tested and sensitivity and specificity were 41.6 and 94.1% respectively. Within 9 hours, 224 patients had two troponin I values and the sensitivity increased to 85.9% with a specificity of 85.0%. At 14 to 24 hours, only 111 patients had three troponin I values and the sensitivity and specificity were 83.6 and 90.1%, respectively. Myoglobin was only measured on admission and was available for 493 of the 528 patients. The sensitivity was 48.0% and specificity was 95.9%.
      Table 3Sensitivity and Specificity of Troponin I and Myoglobin
      MarkerTime 0Up to 9 hUp to 14–24 h
      Troponin I
       Positive, n988450
       Total troponin I tested, n513224111
       Sensitivity41.6%85.9%89.5%
       Specificity94.1%85.0%78.1%
       Positive predictive value80.6%72.6%68.0%
       Negative predictive value96.5%92.9%93.4%
      Myoglobin
      Myoglobin data was available only from the initial blood draw.
       Positive, n98
       Total myoglobin tested, n493
       Sensitivity48.0%
       Specificity95.9%
       Positive predictive value86.7%
       Negative predictive value76.7%
      a Myoglobin data was available only from the initial blood draw.

      Discussion

      This single center, retrospective study of 528 patients demonstrated that safe and effective early diagnosis of myocardial infarction with CK-MB can be achieved in chest pain patients within 9 hours of admission with a sensitivity of 97.4% and a negative predictive value of 98.5%. Only 5 patients (0.9%) had conflicting results, while 523 patients (99.1%) had identical early and final diagnostic outcomes. When admitted from the emergency department, the prevailing hospital protocol for such patients required serial assessment of serum cardiac markers at approximately 0, 8, and 16 hours from the time of admission. For the purpose of our retrospective analysis, an early diagnosis was arbitrarily defined to occur within the first 9 hours of hospitalization. Since most early values actually were available by 8 hours, and many in less than 8 hours, it is entirely possible that similar diagnostic sensitivity might be demonstrated at even earlier time points.
      For the 5 patients in this study who had an apparent delayed rise in CK-MB (after 9 hours), post hoc analysis discloses that in 3 cases there were in-hospital events that might be held accountable. In one instance, the laboratory value is likely to be erroneous. If one incorporates these assumptions, the sensitivity of the test at 9 hours for an out-of-hospital cardiac event may be closer to 99%. One might therefore conclude that, although patients with recurrent chest pain or other in-hospital events require more protracted monitoring and in-hospital care, the evaluation of patients without these events can be safely and effectively completed within 9 hours of presentation to the hospital.
      A recent study in the Netherlands using serial CK-MB mass measurements found the sensitivity to be 40% on admission, 94% at 7 hours, and 97% at 10 hours.
      • de Winter R.J.
      • Bholasingh R.
      • Nieuwenhuijs A.B.
      • et al.
      Ruling out acute myocardial infarction early with two serial creatine kinase-MB mass determinations.
      This retrospective study of 470 patients with an admission diagnosis of typical ischemic chest pain used the World Health Organization (WHO) criteria, in concert with the patient’s clinical status at the time of hospital discharge, to establish a final diagnosis of myocardial infarction. The WHO criteria require the presence of at least 2 of the 3 following findings: clinical history of typical chest pain, evidence of ischemic electrocardiographic changes (ST-segment depression or elevation or T-wave inversions), and a rise of CK-MB mass.

      World Health Organization Criteria for the Diagnosis of Acute Myocardial Infarction. Proposal for the multinational monitoring of trends and determinants in cardiovascular disease Geneva: Cardiovascular Disease Unit of WHO; 2081.

      In the present study, in contrast, enzymatic criteria alone were used to establish the diagnosis of myocardial infarction. The focus on the results of serum enzyme analysis is supported by research demonstrating that, in patients without the classic criteria for myocardial infarction, the presence of elevations of serum CK-MB increases the risk of coronary death and provides additional prognostic value.
      • Pettersson T.
      • Ohlsson O.
      • Tryding N.
      Increased CKMB (mass concentration) in patients without traditional evidence of acute myocardial infarction. A risk indicator of coronary death.
      In addition, in current clinical practice, the third WHO criteria of an elevated CK-MB must be present even if the first two are met.
      The enzymatic diagnosis of myocardial infarction has been extensively evaluated, but most studies have emphasized correctly establishing the diagnosis of myocardial infarction, which requires a high degree of specificity.
      • Adams J.E.
      • Abendschein D.R.
      • Jaffe A.S.
      Biochemical markers of myocardial injury. Is MB creatine kinase the choice for the 1990s?.
      Fewer studies have focused on the exclusion of the myocardial infarction diagnosis, which requires a high sensitivity. Sensitivity should improve if criteria are employed that include all patients with elevated CK-MB, rather than relying solely upon the WHO definition of myocardial infarction. Because the consequences of missed diagnosis are severe, both medically and legally, methods used to diagnose myocardial infarction must possess high sensitivity.
      Another recent study, the Diagnostic Marker Cooperative Study, a prospective analysis performed at 4 hospitals in Houston, Texas, used a diagnostic standard for myocardial infarction based solely upon elevation of CK-MB mass, in an attempt to focus on early exclusion of myocardial infarction.
      • Zimmerman J.
      • Fromm R.
      • Meyer D.
      • et al.
      Diagnostic marker cooperative study for the diagnosis of myocardial infarction.
      In 955 patients, the sensitivity of CK-MB at 6 and 10 hours from onset of chest pain was 66 and 90.4%, respectively. The authors performed their investigation with reference to the time of onset of chest pain; in contrast, the current study was performed with reference to the first phlebotomy. Thus, comparison of the two studies is not readily achievable. It can be conjectured that patients’ retrospective estimates of chest pain duration may be inaccurate and, furthermore, inherent disparities are difficult to document. Therefore, we propose that a more reliable analysis will be obtained using the time of the first blood draw or the documented time of patient arrival, rather than the time of chest pain onset.
      Other studies have also evaluated timing using CK-MB. An examination of 100 patients demonstrated a 95% sensitivity for CK-MB at 8 hours from admission with myocardial infarction diagnosis, based upon WHO criteria.
      • Mohler E.R.
      • Ryan T.
      • Segar D.S.
      • et al.
      Clinical utility of troponin T levels and echocardiography in the emergency department.
      Another study examined 309 patients, using similar composite clinical and enzymatic criteria for diagnosis, and showed sensitivity for CK-MB mass to be 95% at 8 hours and 99% at 12 hours from the time of chest pain onset.
      • de Winter R.J.
      • Koster R.W.
      • Sturk A.
      • et al.
      Value of myoglobin, troponin T, and CK-MB mass in ruling out an acute myocardial infarction in the emergency room.
      In an earlier study involving 616 patients and using WHO criteria, sensitivity for CK-MB was 90% at 7 hours, and 100% at 11 hours of chest pain duration.
      • Gibler W.B.
      • Young G.P.
      • Hedges J.R.
      • et al.
      Acute myocardial infarction in chest pain patients with nondiagnostic ECGs: serial CK-MB sampling in the emergency department. The Emergency Medicine Cardiac Research Group.
      This data again lends support to the conclusion that early diagnosis can be achieved with high sensitivity. However, as previously discussed, studies that use the WHO criteria may not optimize diagnostic sensitivity.
      Other markers have also been evaluated for early diagnosis. Several older studies suggested that myoglobin might be useful for the very early evaluation of chest pain patients
      • Gibler W.B.
      • Gibler C.D.
      • Weinshenker E.
      • et al.
      Myoglobin as an early indicator of acute myocardial infarction.
      • Tucker J.F.
      • Collins R.A.
      • Anderson A.J.
      • et al.
      Value of serial myoglobin levels in the early diagnosis of patients admitted for acute myocardial infarction.
      • Montague C.
      • Kircher T.
      Myoglobin in the early evaluation of acute chest pain.
      ; more recently, interest in myoglobin has diminished because of the concern over low specificity for cardiac damage.
      • Falahati A.
      • Sharkey S.W.
      • Christensen D.
      • et al.
      Implementation of serum cardiac troponin I as marker for detection of acute myocardial infarction.
      Most current investigations focus on cardiac troponin I, troponin T, or both. Although much of the research has featured the prognostic role of troponin in patients with unstable angina, some studies have shown that troponin may be at least as effective as CK-MB for early diagnosis, if not superior. A study of 773 chest pain patients demonstrated a 100% sensitivity of troponin I at 4 hours after admission.
      • Hamm C.W.
      • Goldmann B.U.
      • Heeschen C.
      • et al.
      Emergency room triage of patients with acute chest pain by means of rapid testing for cardiac troponin T or troponin I.
      Another study showed that troponin T achieved 100% sensitivity after 8 hours.
      • Mohler E.R.
      • Ryan T.
      • Segar D.S.
      • et al.
      Clinical utility of troponin T levels and echocardiography in the emergency department.
      In other hands, the results have not been quite as promising.
      • Falahati A.
      • Sharkey S.W.
      • Christensen D.
      • et al.
      Implementation of serum cardiac troponin I as marker for detection of acute myocardial infarction.
      • Antman E.M.
      • Grudzien C.
      • Sacks D.B.
      Evaluation of a rapid bedside assay for detection of serum cardiac troponin T.
      The results presented here must be evaluated in the appropriate context. Negative cardiac markers, regardless of the time period over which they are collected, do not rule out the presence of ischemic coronary disease. Negative enzyme testing indicates that there has been no detectable cardiac damage. A patient may still have unstable angina and be at risk for future events and, therefore, require further work-up. Clinical acumen must determine what further evaluation is required for each patient. A rapid rule-out protocol moves up this decision point but does not eliminate it.
      It is also important to consider the population to which these results apply. The patients were all seen at Stanford University Hospital, a large academic center in a suburban location. The length of time between chest pain onset and arrival at the hospital, which was not determined in the study, has an effect on the sensitivity of cardiac enzymes at different time points. Population characteristics and local emergency services influence this length of time. In addition, the study looked only at patients who were admitted to the hospital and the threshold for admission in the emergency department may differ from other facilities. Finally, at least three enzyme measurements at specific time points were required for inclusion in the study; therefore, 335 of the 863 patients admitted through the emergency department were not evaluated. The impact of these factors is unknown. Despite some potential limitations, the study does provide useful information if applied to the appropriate population.
      In summary, standard CK-MB mass measurements within 9 hours of the patient’s hospital arrival provided an accurate, objective diagnostic assessment. The high sensitivity and negative predictive values suggest that early enzymatic diagnosis of myocardial infarction is feasible and reliable. Although other studies have evaluated the time-dependent diagnostic accuracy of CK-MB, we feel that our study design, using an enzymatic diagnosis of myocardial infarction and assessment linked to the timing of the first phlebotomy, sheds new light on the early diagnostic approach to myocardial necrosis. Improving the efficiency of the early evaluation of chest pain would permit earlier intervention and treatment and might provide substantial cost savings. In addition, decreasing the time and expense of the evaluation may decrease the disincentive to observe patients with apparent low-risk and might thereby reduce the number of neglected myocardial infarctions. Some studies of chest pain observation units using accelerated diagnostic protocols have demonstrated cost savings and reduction in missed myocardial infarctions,
      • Roberts R.R.
      • Zalenski R.J.
      • Mensah E.K.
      • et al.
      Costs of an emergency department-based accelerated diagnostic protocol vs hospitalization in patients with chest pain: a randomized controlled trial.
      • Graff L.G.
      • Dallara J.
      • Ross M.A.
      • et al.
      Impact on the care of the emergency department chest pain patient from the chest pain evaluation registry (CHEPER) study.
      but these studies have predominantly included only low-risk patients. The results of this retrospective study of chest pain, performed in patients from all risk categories, strongly suggest that an early diagnosis protocol would merit evaluation in a prospective study and that it is likely to be successful.

      References

        • Nourjah P.
        National Hospital Ambulatory Medical Care Survey: 1997 emergency department summary.
        Adv Data. 1999; 304: 1-24
      1. The DRG handbook: comparative clinical and financial benchmarks. HCIA Inc, Baltimore1999
        • Barish R.A.
        • Doherty R.J.
        • Browne B.J.
        Reengineering the emergency evaluation of chest pain.
        J Healthc Qual. 1997; 19: 6-12
        • Gibler W.B.
        • Young G.P.
        • Hedges J.R.
        • et al.
        Acute myocardial infarction in chest pain patients with nondiagnostic ECGs: serial CK-MB sampling in the emergency department. The Emergency Medicine Cardiac Research Group.
        Ann Emerg Med. 1992; 21: 504-512
        • Rude R.E.
        • Poole W.K.
        • Muller J.E.
        • et al.
        Electrocardiographic and clinical criteria for recognition of acute myocardial infarction based on analysis of 3,697 patients.
        Am J Cardiol. 1983; 52: 936-942
        • Behar S.
        • Schor S.
        • Kariv I.
        • et al.
        Evaluation of electrocardiogram in emergency room as a decision-making tool.
        Chest. 1977; 71: 486-491
        • McCarthy B.D.
        • Beshansky J.R.
        • D’Agostino R.B.
        • et al.
        Missed diagnoses of acute myocardial infarction in the emergency department: results from a multicenter study.
        Ann Emerg Med. 1993; 22: 579-582
        • Schor S.
        • Behar S.
        • Modan B.
        • et al.
        Disposition of presumed coronary patients from an emergency room. A follow-up study.
        JAMA. 1976; 236: 941-943
        • Sigurdsson E.
        • Thorgeirsson G.
        • Sigvaldason H.
        • et al.
        Unrecognized myocardial infarction: epidemiology, clinical characteristics, and the prognostic role of angina pectoris. The Reykjavik Study.
        Ann Intern Med. 1995; 122: 96-102
        • Lee T.H.
        • Rouan G.W.
        • Weisberg M.C.
        • et al.
        Clinical characteristics and natural history of patients with acute myocardial infarction sent home from the emergency room.
        Am J Cardiol. 1987; 60: 219-224
        • de Winter R.J.
        • Bholasingh R.
        • Nieuwenhuijs A.B.
        • et al.
        Ruling out acute myocardial infarction early with two serial creatine kinase-MB mass determinations.
        Eur Heart J. 1999; 20: 967-972
      2. World Health Organization Criteria for the Diagnosis of Acute Myocardial Infarction. Proposal for the multinational monitoring of trends and determinants in cardiovascular disease Geneva: Cardiovascular Disease Unit of WHO; 2081.

        • Pettersson T.
        • Ohlsson O.
        • Tryding N.
        Increased CKMB (mass concentration) in patients without traditional evidence of acute myocardial infarction. A risk indicator of coronary death.
        Eur Heart J. 1992; 13: 1387-1392
        • Adams J.E.
        • Abendschein D.R.
        • Jaffe A.S.
        Biochemical markers of myocardial injury. Is MB creatine kinase the choice for the 1990s?.
        Circulation. 1993; 88: 750-763
        • Zimmerman J.
        • Fromm R.
        • Meyer D.
        • et al.
        Diagnostic marker cooperative study for the diagnosis of myocardial infarction.
        Circulation. 1999; 99: 1671-1677
        • Mohler E.R.
        • Ryan T.
        • Segar D.S.
        • et al.
        Clinical utility of troponin T levels and echocardiography in the emergency department.
        Am Heart J. 1998; 135: 253-260
        • de Winter R.J.
        • Koster R.W.
        • Sturk A.
        • et al.
        Value of myoglobin, troponin T, and CK-MB mass in ruling out an acute myocardial infarction in the emergency room.
        Circulation. 1995; 92: 3401-3407
        • Gibler W.B.
        • Gibler C.D.
        • Weinshenker E.
        • et al.
        Myoglobin as an early indicator of acute myocardial infarction.
        Ann Emerg Med. 1987; 16: 851-856
        • Tucker J.F.
        • Collins R.A.
        • Anderson A.J.
        • et al.
        Value of serial myoglobin levels in the early diagnosis of patients admitted for acute myocardial infarction.
        Ann Emerg Med. 1994; 24: 704-708
        • Montague C.
        • Kircher T.
        Myoglobin in the early evaluation of acute chest pain.
        Am J Clin Pathol. 1995; 104: 472-476
        • Falahati A.
        • Sharkey S.W.
        • Christensen D.
        • et al.
        Implementation of serum cardiac troponin I as marker for detection of acute myocardial infarction.
        Am Heart J. 1999; 137: 332-337
        • Hamm C.W.
        • Goldmann B.U.
        • Heeschen C.
        • et al.
        Emergency room triage of patients with acute chest pain by means of rapid testing for cardiac troponin T or troponin I.
        N Engl J Med. 1997; 337: 1648-1653
        • Antman E.M.
        • Grudzien C.
        • Sacks D.B.
        Evaluation of a rapid bedside assay for detection of serum cardiac troponin T.
        JAMA. 1995; 273: 1279-1282
        • Roberts R.R.
        • Zalenski R.J.
        • Mensah E.K.
        • et al.
        Costs of an emergency department-based accelerated diagnostic protocol vs hospitalization in patients with chest pain: a randomized controlled trial.
        JAMA. 1997; 278: 1670-1676
        • Graff L.G.
        • Dallara J.
        • Ross M.A.
        • et al.
        Impact on the care of the emergency department chest pain patient from the chest pain evaluation registry (CHEPER) study.
        Am J Cardiol. 1997; 80: 563-568