Original Contribution

The risk for acute coronary syndrome associated with atrial fibrillation among ED patients with chest pain syndromes

Aaron M. Brown BS, Keara L. Sease MAEd, Jennifer L. Robey MSN, BSN, Frances S. Shofer PhD, Judd E. Hollander MD*

Department of Emergency Medicine, Hospital of the University of Pennsylvania, Philadelphia, PA 19104, USA

Received 18 August 2006; revised 15 September 2006; accepted 29 September 2006

Abstract

Objective: We sought to determine if atrial fibrillation is associated with an increased risk for an acute coronary syndrome (ACS) among emergency department (ED) patients with chest pain syndromes. Methods: We performed a retrospective analysis of a prospectively collected database on ED patients with chest pain by selecting patients with atrial fibrillation and frequency-matched Control subjects without atrial fibrillation. Measured outcomes were acute myocardial infarction (AMI), ACS, and unstable angina . The Relative risks of AMI, ACS, and UA associated with atrial fibrillation were calculated.

Results: One hundred forty patients with atrial fibrillation and 683 matched control subjects were identified. The rates of AMI for the atrial fibrillation and control groups were 11.4% and 10.8%, respectively; those of ACS were 27.9% and 26.7%, respectively; and those of UAwere 16.4% and 15.8%, respectively. The relative risks of AMI and ACS did not increase in patients with atrial fibrillation: AMI, 1.05 (95% confidence interval [CI] = 0.63-1.75); ACS, 1.05 (95% CI = 0.78-1.40); and UA, 1.05

(95% CI = 0.6-1.7).

Conclusion: Among patients presenting to the ED with chest pain syndromes, atrial fibrillation is not associated with an increased risk for AMI, ACS, and UA.

D 2007

Introduction

Atrial fibrillation is the most common cardiac dysrhyth- mia, accounting for more than 2.3 million cases in the United States [1]. One percent of all patients presenting to the emergency department (ED) have atrial fibrillation [2], and the risk for myocardial infarction among these patients is 5.5% [3]. new-onset atrial fibrillation is known to occur in the setting of an acute myocardial infarction (AMI) [4-7].

* Corresponding author. Tel.: +1 215 662 6698; fax: +1 215 662 3953.

In addition, atrial fibrillation is associated with increased in- hospital mortality in the setting of an acute coronary syndrome (ACS) [8-10]. Approximately 39% of patients with atrial fibrillation present with concurrent chest pain [3]. Among ED patients with atrial fibrillation, lack of chest pain has been associated with a lower risk for AMI, suggesting that lone atrial fibrillation in patients without Traditional risk factors or a presentation otherwise suggestive of ACS may not require evaluation to rule out ACS [3,11]. Many risk factors for ACS have been examined among the approxi- mately 6 million patients who present with chest pain to EDs in the United States each year [12-19]. However, it is not known if atrial fibrillation by itself is a risk factor for

0735-6757/$ – see front matter D 2007 doi:10.1016/j.ajem.2006.09.015

myocardial ischemia among patients presenting to the ED with chest pain suspected to be associated with ACS or simply a marker of previously identified risk factors among these patients.

We hypothesized that the presence of atrial fibrillation was not a risk factor for AMI or ACS among patients presenting to the ED with chest pain suspected to be associated with a potential ACS.

Methods

Study design

We performed a retrospective analysis of a prospectively collected cohort of ED patients with chest pain comparing matched cohorts of patients with and those without atrial fibrillation to determine whether the presence of atrial fibrillation upon ED arrival is associated with AMI, ACS, or Unstable angina . Frequency matching was used to select a control cohort with similar baseline characteristics and cardiac risk factors.

Setting

This study was conducted at the ED of an urban tertiary care center with an annual census of 55000 patients and an annual census of approximately 2000 patients with chest pain.

Selection of participants

Patients were selected from 2 prospectively collected cohorts of ED patients with chest pain syndromes. The first group was collected as part of an observational quality assurance protocol of consecutive ED patients aged 18 years or older with chest pain syndromes (chest pain, dyspnea, syncope) for the larger Internet Tracking Registry for Acute Coronary Syndromes from July 1999 through August 2001 [20]. The second group was a convenience sample of patients with chest pain syndromes prompting an electro- cardiogram (ECG) who consented to be enrolled in biochemical marker studies from 2002 through October 2004 at a single institution [19]. Both groups had a follow- up rate greater than 97% completed at 30 days. The patients with atrial fibrillation and matched control subjects were retrospectively selected from these prospectively collected cohorts. During the study periods, trained research assistants were present 16 hours a day and 7 days a week to identify and enroll potential patients in both groups. All subjects included in the biochemical marker studies provided written consent, and enrollment in any of the protocols did not alter the treatment of any patient.

From these cohorts, all patients with chest pain syn- dromes who were at least 25 years old were eligible for inclusion in this study. Patients with a self-reported history of cocaine use within the last week or a positive urine drug screen finding for cocaine were excluded. The atrial

fibrillation group was identified based on the rhythm interpretation of the treating physician and matched with control subjects without atrial fibrillation. To reduce confounding, we used frequency matching to increase the likelihood of an even distribution of historical baseline characteristics known to predict acute myocardial ischemia between the 2 populations. Patients with atrial fibrillation with multiple visits during the data collection period had only their First visit selected, and control subjects with multiple visits could have only one of their visits randomly selected. The atrial fibrillation population was stratified based on age categories (b51, 51-60, 61-70, 71-80, and N80 years), sex, race (black and non-black), and reported history of coronary artery disease (CAD) or AMI. Based on the number of patients in each stratum among the atrial fibrillation population, the control subjects were randomly selected from an equivalent stratum. Random selection was used to minimize confounding from nonmatched variables, including onset, duration and character of chest pain, associated symptoms, cardiac risk factors, ECG changes, biochemical marker elevation, and hemodynamics. Specific variables (eg, shortness of breath and palpitations) were not matched on to prevent overmatching of factors possibly related to atrial fibrillation that would create exposed and unexposed populations that were too homogeneous and result in bias toward a null result. The studies from which data were collected were approved by the academic center’s institutional review board.

Methods of measurement

Patients who were enrolled in the databases had a structured history and physical examination performed for them at the time of presentation. Patient information collected consisted of demographic characteristics, pertinent medications, chest pain characteristics, associated symp- toms, cardiac risk factors, pertinent past medical conditions, prior cardiac testing, treatment in the ED, test results, and disposition in accordance to Standardized reporting guide- lines [21]. Patient data were prospectively obtained from the treating physician in a closed-question format. Hospital course was prospectively followed daily to reliably record patient complications, interventions, laboratory results, and results of diagnostic testing.

All initial ECGs were evaluated by the treating physician and examined for the rhythm and presence or absence of ST elevation, ST depression, T-wave inversions, pathologic Q waves, Left bundle-branch block, and right bundle-branch block as well as to determine if the finding was known to be old.

Outcome measures

The main outcomes examined were AMI, ACS, and UA. Acute coronary syndrome was defined as AMI and UA. For each visit, the final diagnosis at discharge was classified as AMI, UA, nonischemic chest pain, or chest pain not

otherwise specified. The outcomes were all based on the final hospital diagnosis from the index visit and defined in accordance to standardized reporting guidelines [21]. Acute myocardial infarction was defined as a creatine kinase MB (CK-MB) level of 10 ng/mL or higher (N2 times higher than normal; Abbott Axsym CK-MB, Abbott, Abbott Park, Ill) or a troponin I level greater than 2 ng/mL (normal = b0.4 ng/mL; indeterminate = 0.4-2 ng/mL; Abbot Axsym Troponin I) with or without definitive changes on ECG in accordance to European Society of Cardiology/American College of Cardiology guidelines [22].

Unstable angina was defined in accordance to standard- ized reporting guidelines as referring those patients with consistent symptoms and proven underlying coronary disease, including a positive exercise stress test (ST-segment depression = N1.5 mm), reversible Ischemic changes on stress echocardiogram or sestamibi scan, blockage of at least one coronary artery on angiography of 70% or greater, or elevation of troponin I or CK-MB at a level higher than normal but lower than the limits set for AMI (0.3 ng/mL b troponin I V 2 ng/mL or 5 ng/mL V CK-MB b 10 ng/mL) [21,23]. Although this definition equates documented CAD with UA, we assumed that each patient received a diagnostic test owing to concerns about the possibility of having UA

Table 1 Baseline characteristics of the atrial fibrillation and matched control groups

^a Most recent or current episode of chest pain.

^b Most recent or current episode of chest pain.

and therefore considered the demonstration of reversible ischemia or CAD as a confirmation of UA.

Patients without significant CAD on cardiac catheteriza- tion or reversible ischemia on stress imaging modality were classified as nonischemic. Patients with an alternate diagnosis confirmed with definitive testing were also classified as nonischemic if there was a negative ischemia workup. All other patients, including those who did not receive an objective evaluation, were classified as having chest pain not otherwise specified.

Data analysis

Data were entered into a Microsoft Access 2000 database (Microsoft Corporation, Redmond, Wash) and imported into a SAS statistics software (version 9.1, SAS Institute, Cary, NC). The baseline characteristics of the patients with and those without atrial fibrillation were reported in accordance to standardized reporting guidelines [21]. Continuous variables were represented as mean values with standard deviations or as median values with interquartile ranges. Discrete variables were represented as frequencies with percentages. Baseline characteristics were compared between the groups using Fisher’s exact test for discrete variables, Student’s t test for

Patient characteristic	Atrial fibrillation group (n = 140)	Control group (n = 683)	P
Age (y; mean F SD)	66.4 F 12.9	65.4 F 13.5	.44
Male sex [n (%)]	71 (50.7)	342 (50.1)	.92
Race [n (%)]
Black	72 (51.4)	357 (52.3)	.92
Other	68 (48.6)	326 (47.7)
Cardiac risk factors [n (%)]
Hypertension	85 (60.7)	413 (60.5)	1.0
Family history of AMI	26 (18.6)	109 (16.0)	.45
Diabetes mellitus	29 (20.7)	171 (25.0)	.33
Tobacco use	45 (32.1)	246 (36.0)	.44
Hyperlipidemia	36 (25.7)	180 (26.4)	.92
Past cardiac events [n (%)]
Congestive heart failure	38 (27.1)	166 (24.3)	.52
AMI	31 (23.0)	125 (18.6)	.23
known CAD	45 (32.1)	227 (33.2)	.84
Characteristics of chest pain [mean (interquartile range)]
Duration of symptoms (min)a	120 (20-300)	90 (15-480)	.89
Time since onset of chest pain (min)b	180 (60-1440)	240 (60-1440)	.78
Associated symptoms [n (%)]
Dyspnea	91 (65.0)	327 (48.0)	.0003
Diaphoresis	35 (25.0)	147 (21.6)	.37
Nausea/vomiting	15 (10.7)	38 (5.6)	.04
Palpitations	47 (33.6)	61 (9.0)	b.0001
Vital signs (mean F SD) Systolic blood pressure (mm Hg)	129 F 28.1	145 F 26.1	b.0001
Diastolic blood pressure (mm Hg)	78.9 F 18.2	80.8 F 16.1	.14
Pulse rate (beats/min)	100 F 27.8	83.6 F 17.6	b.0001

	Table 2 Changes in ECGa not known to be old for the 2 groups Atrial fibrillation Control P group group					Table 4 Relative risks of outcomes for the atrial fibrillation group as compared with the control group RR (95% CI) AMI 1.05 (0.63-1.75)
		(n = 140)	(n = 683)			ACS	1.05 (0.78-1.40)
	ST elevation	10 (7.1)	47 (6.9)	.86		UA	1.05 (0.6-1.7)
	ST depression	22 (15.7)	49 (7.2)	.003		Death	1.5 (0.35-5.0)
	T-wave inversion	37 (26.4)	158 (23.2)	.45
	Left bundle-branch	3 (2.1)	10 (1.5)	.47

block

Right bundle-branch block

2 (1.4) 11 (1.6) 1.0

out atrial fibrillation were selected via frequency matching as control subjects.

Q waves 4 (2.9) 22 (3.2) 1.0

Data are expressed as n (%).

^a As read by the treating physician.

continuous variables, and the Wilcoxon rank sum test for nonparametric continuous statistics.

The rates of AMI and ACS among the patients with and those without atrial fibrillation were calculated and com- pared using Fisher’s exact test. Relative risks (RRs) were calculated for each outcome with 95% confidence intervals (CIs) to compare the 2 populations. Ninety-five percent CIs were calculated in StatXact 6.1 (Cytel Software Corpora- tion, Cambridge, Mass).

Results

Population characteristics

Five thousand five hundred fifty-seven eligible patients with chest pain and ECGs were included in the 2 databases. From this population, there were 4715 unique patients without atrial fibrillation and 140 unique patients with atrial fibrillation. Six hundred eighty-three unique patients with-

Table 3 Outcomes for the atrial fibrillation and matched

control groups

Atrial fibrillation Control group group (n = 140) (n = 683)

There was no statistical difference between the groups.

	n (%)	95% CI		n (%)	95% CI
AMI	16 (11.4)	6.7-17.9	74 (10.8)		8.6-13.4
UA	23 (16.4)	10.7-23.6	108 (15.8)		13.2-18.8
ACS	39 (27.9)	20.6-36.1	182 (26.7)		23.4-16.1
Nonischemic	17 (12.1)	7.2-18.7	91 (13.3)		10.9-16.1
chest pain
Chest pain not	81 (57.9)	49.2-66.2	387 (56.7)		52.9-60.4
otherwise
specified
Other definitive	3 (2.1)	0.4-6.1	23 (3.4)		2.2-5.0
diagnoses
Death	4 (2.9)	0.8-7.2	13 (1.9)		1.0-3.3

There was no significant difference between the groups

on the matched baseline characteristics. The populations were similar with respect to age, race, sex, history of CAD, and history of AMI as well as other historical factors other than dyspnea and palpitations, which are both symptoms of atrial fibrillation (Table 1). The 2 groups were similar with respect to rates of new ST elevations, T-wave inversions, Q waves, left bundle-branch block, and right bundle-branch block (Table 2). However, patients with atrial fibrillation had a higher mean pulse rate (100 vs 84 beats/min) as well as a lower mean systolic blood pressure (135 vs 145 mm Hg) and were more likely to have new ST depressions (16% vs 7%) (Tables 1 and 2). Other demographic data, associated symptoms, pertinent medical history, and ECG changes are depicted in Tables 1 and 2.

Relationship between atrial fibrillation and patient outcomes

Final hospital diagnoses were similar among the patients with atrial fibrillation and the control subjects (Tables 3 and 4). Atrial fibrillation did not increase the likelihood of AMI (RR = 1.05; 95% CI = 0.63-1.75), ACS (RR = 1.05; 95% CI = 0.78-1.40), and UA (RR = 1.05; 95% CI = 0.6-1.7).

Discussion

Previous studies have reported a relatively low risk ranging from 2% to 5% for myocardial ischemia among ED patients with atrial fibrillation and established that the traditional predictors of ischemia hold true for this popula- tion, including typical chest pain and ST-segment deviation [3,11]. These studies included patients with atrial fibrillation only, so they were unable to elucidate if the presence of atrial fibrillation was associated with additional risk for ACS among ED patients with potential ACS (or chest pain suspected to be associated with ACS). We have demonstrat- ed that atrial fibrillation does not increase the risk for AMI or ACS among a cohort of ED patients with chest pain suspected to be associated with potential ACS.

A link between myocardial ischemia and atrial fibrillation has been proposed based on multiple studies on patients after AMI and patients with previously documented CAD [4-7,9,10]. We confirmed that patients presenting with both

chest pain and atrial fibrillation are a high-risk population, with approximately 36% having a history of CAD or AMI, most having multiple cardiac risk factors and a mean age of 65 years, and 28% presenting with ACS. Although myocar- dial ischemia and myocardial injury may be triggers for the aberrant conduction, the presence of atrial fibrillation was not a predictor of myocardial ischemia in the acute setting among ED patients with chest pain syndromes. Atrial fibrillation may simply reflect myocardial disease significant enough to cause dysrhythmia, which is already accounted for by other historical factors, such as CAD and prior AMI. In addition, factors specific to the acute care visit, such as typical chest pain and ECG changes, are stronger predictors of an acute ischemic event in the ED as compared with historical coronary risk factors [13,24-28]. Atrial fibrillation is prob- ably more akin to a historical factor reflecting past disease rather than a sign of acute Plaque rupture or tissue ischemia. In our study, patients in the atrial fibrillation and control groups were well matched on cardiac risk factors. However, patients with atrial fibrillation did differ on most presenting characteristics; they were more likely to have dyspnea, palpitations, and ECG depressions and had a lower mean systolic blood pressure as well as a higher ventricular rate. These differences are not surprising considering the decreased cardiac output and increased rate associated with atrial fibrillation. Many of these factors have been associ- ated with a higher likelihood of acute ischemia among patients with chest pain [13,25,26], but the rates of acute ischemia among the 2 groups in this study were similar. These suggest that the differences in the presenting characteristics were related to the presence of atrial

fibrillation and not an acute ischemic event.

The presence of a difference in factors related to atrial fibrillation suggests that overmatching did not occur and that overmatching likely cannot explain the similar rates of acute ischemia in the 2 groups. Atrial fibrillation appears to be a proxy for a population with chronic heart disease; when this population is matched with a population with similar historical cardiac risk factors and Disease burden, atrial fibrillation does not increase the risk for ACS. We could have adjusted for the factors specific to the index visit that were different between the 2 groups; however, as previously stated, they seem to be atrial fibrillation related and adjusting based on them would present bias toward the null and not change the main conclusion of the study.

Despite the fact that the presence of atrial fibrillation alone does not predict an increased risk for ACS among patients with chest pain syndromes, patients with chest pain syndromes and atrial fibrillation are still at high risk for ACS (28%). Although admission to rule out ischemia may not be justified in all patients with atrial fibrillation seen in the ED, admission to rule out ischemia is justified among ED patients with atrial fibrillation when they have related chest pain symptoms that are suspected to be associated with ACS. In addition, patients with new-onset atrial fibrillation in the setting of ACS have higher morbidity

and mortality [4-10]; thus, aggressive treatment should be pursued if ACS is suspected on grounds other than simply the presence of atrial fibrillation.

Limitations

We acknowledge several limitations of this study. This was a retrospective analysis of a prospectively collected database that used matching–it has the limitations inherent in this type of study. Selection bias was limited by screening all patients presenting to the ED with chest pain during the enrollment periods. All patients with atrial fibrillation were included, and random samplings of patients without atrial fibrillation from a matched stratum were selected to limit unknown confounders. The lack of Definitive imaging testing for all patients could lead to either the overdiagnosis or the underdiagnosis of ACS; however, this was an observational study, so testing was not conducted unless clinically indicated. We did not distinguish between new- onset atrial fibrillation and chronic atrial fibrillation; neither did we adjust for heart rate at the time of presentation. Patients with chronic atrial fibrillation would be less likely to have acute changes in myocardial oxygen Supply and demand leading to ischemia secondary to atrial fibrillation. However, most of the patients with demand ischemia in our study would have undergone definitive testing to show that the biochemical marker elevation was a result of demand and not an underlying coronary disease. In the end, this would not have changed the rates of our outcomes. As previously mentioned, not all factors were controlled for, including those specific to the presentation, such as ECG changes and chest pain characteristics.

Conclusions

Atrial fibrillation is not associated with an increased risk for AMI or ACS among patients presenting to the ED with chest pain syndromes. Therefore, dispositions of decisions regarding ruling out acute ischemia among patients with chest pain syndromes should not be altered by the presence of atrial fibrillation and should instead be based on the presence of other previously identified risk factors.

References

1. Go AS, Hylek EM, Phillips KA, et al. Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and Stroke prevention: the AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) study. JAMA 2001;285:2370 – 5.
2. Scott PA, Pancioli AM, Davis LA, et al. Prevalence of atrial fibrillation and antithrombotic prophylaxis in emergency department patients. Stroke 2002;33:2664 – 9.
3. Zimetbaum PJ, Josephson ME, McDonald MJ, et al. Incidence and predictors of myocardial infarction among patients with atrial fibrillation. J Am Coll Cardiol 2000;36:1223 – 7.
4. Liberthson RR, Salisbury KW, Hutter Jr AM, et al. Atrial tachyar- rhythmias in acute myocardial infarction. Am J Med 1976;60:956 – 60.
5. Crenshaw BS, Ward SR, Granger CB, et al. Atrial fibrillation in the setting of acute myocardial infarction: the GUSTO-I experience. Global Utilization of Streptokinase and TPA for Occluded Coronary Arteries. J Am Coll Cardiol 1997;30:406 – 13.
6. Wong CK, White HD, Wilcox RG, et al. New atrial fibrillation after acute myocardial infarction independently predicts death: the GUS- TO-III experience. Am Heart J 2000;140:878 – 85.
7. Eldar M, Canetti M, Rotstein Z, et al. Significance of Paroxysmal atrial fibrillation complicating acute myocardial infarction in the thrombolytic era. Circulation 1998;97:965 – 70.
8. Mehta RH, Dabbous OH, Granger CB, GRACE Investigators, et al. Comparison of outcomes of patients with acute coronary syndromes with and without atrial fibrillation. Am J Cardiol 2003;92:1031 – 6.
9. Cameron A, Schwartz MJ, Kronmal RA, Kosinski AS, et al. Prevalence and significance of atrial fibrillation in coronary artery disease (CASS Registry). Am J Cardiol 1988;61:714 – 7.
10. Pizzetti F, Turazza FM, Franzosi MG, GISSI-3 Investigators, et al. Incidence and prognostic significance of atrial fibrillation in acute myocardial infarction: the GISSI-3 data. Heart 2001;86:527 – 32.
11. Friedman HZ, Weber-Bornstein N, Deboe SF, et al. Cardiac care unit admission criteria for suspected acute myocardial infarction in new- onset atrial fibrillation. Am J Cardiol 1987;59:866 – 9.
12. McCaig LF, Burt CW. National Hospital Ambulatory Medical Care Survey: 2002 emergency department summary. Advance data from Vital and Health Statistics, Atlanta, GA, CDC; 2004.
13. Panju AA, Hemmelgarn BR, Guyatt GH, et al. The rational clinical examination. Is this patient having a myocardial infarction? JAMA 1998;280:1256 – 63.
14. Goldman L, Weinberg M, Weisberg M, Olshen R, Cook EF, Sargent RK, et al. A computer-derived protocol to aid in the diagnosis of emergency room patients with acute chest pain. N Engl J Med 1982; 307:588 – 96.
15. Lee TH, Cook EF, Weisberg M, Sargent RK, Wilson C, Goldman L. Acute chest pain in the emergency room. Identification and examination of low-risk patients. Arch Intern Med 1985;145:65 – 9.
16. Pozen MW, D’Agostino RB, Selker HP, Sytkowski PA, Hood Jr WB. A predictive instrument to improve coronary-care-unit admission practices in acute ischemic heart disease. A prospective multicenter clinical trial. N Engl J Med 1984;310:273 – 8.
17. Selker HP, Beshansky JR, Griffith JL, Aufderheide TP, Ballin DS, Bernard SA, et al. Use of the Acute cardiac ischemia Time- Insensitive Predictive Instrument (ACI-TIPI) to assist with triage of patients with chest pain or other symptoms suggestive of acute cardiac ischemia. A multicenter, controlled clinical trial. Ann Intern Med 1998;129:845 – 55.
18. Antman EM, Cohen M, Bernink PJ, McCabe CH, Horacek T, Papuchis G, et al. The TIMI risk score for unstable angina/non-ST elevation MI: a method for prognostication and therapeutic decision making. JAMA 2000;284:835 – 42.
19. Chase M, Zogby K, Shofer FS, Sease K, Robey JL, Hollander JE. Prospective validation of the TIMI risk score in the emergency department chest pain population. Ann Emerg Med 2006;48:252 – 9.
20. Blomkalns AL, Lindsell CJ, Chandra A, Osterlund ME, Gibler WB, Pollack CV, et al. Can electrocardiographic criteria predict adverse cardiac events and positive cardiac markers? Acad Emerg Med 2003; 10:205 – 10.
21. Hollander JE, Blomkalns AL, Brogan GX, Multidisciplinary Stan- dardized Reporting Criteria Task Force, et al. Standardized reporting guidelines for studies evaluating risk stratification of ED patients with potential acute coronary syndromes. Acad Emerg Med 2004;11: 1331 – 40.
22. Alpert JS, Thygesen K, Antman E, et al. Myocardial infarction redefined–a consensus document of the Joint European Society of Cardiology/American College of Cardiology Committee for the Redefinition of Myocardial Infarction. J Am Coll Cardiol 2000;36: 959 – 69.
23. Braunwald E, Antman EM, Beasley JW, American College of Cardiology, American Heart Association, Committee on tha Manage- ment of Patients with Unstable Angina, et al. ACC/AHA 2002 guideline update for the management of patients with unstable angina and Non-ST-segment elevation myocardial infarction–summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on the Management of Patients With Unstable Angina). J Am Coll Cardiol 2002;40:1366 – 74.
24. Jayes Jr RL, Beshansky JR, D’Agostino RB, et al. Do patients’ coronary risk factor reports predict acute cardiac ischemia in the emergency department? A multicenter study. J Clin Epidemiol 1992; 45:621 – 6.
25. Lee TH, Goldman L. Evaluation of the patient with acute chest pain. N Engl J Med 2000;342:1187 – 95.
26. Rouan GW, Lee TH, Cook EF, et al. Clinical characteristics and outcome of acute myocardial infarction in patients with initially normal or nonspecific electrocardiograms. Am J Cardiol 1989;64:1087 – 92.
27. Braunwald E, Jones RH, Mark DB, et al. Unstable angina: diagnosis and management. Clinical practice guideline number 10. Rockville (Md)7 Agency for Health Care Policy and Research and the National Heart, Lung, and Blood Institute, Public Health Service, U.S. Department of Health and Human Services; 1994.
28. Zalenski RJ, Shamsa F, Pede KJ. Evaluation and risk stratification of patients with chest pain in the emergency department. Predictors of life-threatening events. Emerg Med Clin North Am 1998;16:495 – 517.