Original Contribution

Combining Thrombolysis in Myocardial Infarction risk score and clear-cut alternative diagnosis for chest pain risk stratification

Caren F. Campbell, Anna Marie Chang MD?, Keara L. Sease, Christopher Follansbee, Christine M. McCusker RN, Frances S. Shofer PhD, Judd E. Hollander MD

Department of Emergency Medicine, Ground Ravdin, Hospital of the University of Pennsylvania, Philadelphia

Received 8 December 2007; revised 11 January 2008; accepted 12 January 2008

Abstract

Objective: The Thrombolysis in Myocardial Infarction risk score is a validated risk stratification tool useful in patients with definite and potential acute coronary syndromes but does not identify patients safe for discharge from the emergency department (ED). Likewise, the use of a clear-cut alternative Noncardiac diagnosis risk stratifies patients but does not identify a group safe for discharge. We hypothesized that the presence of an alternative diagnosis in patients with a TIMI risk score less than 2 might identify a cohort of patients safe for ED discharge.

Methods: In prospective cohort study, we enrolled ED patients with potential ACS. Data included demographics, medical history, components of the TIMI risk score, and whether the treating physician ascribed the condition to an alternative noncardiac diagnosis. Investigators followed the patients through the hospital course, and 30-day follow-up was done. The main outcome was 30-day death, myocardial infarction, or revascularization.

Results: A total of 3169 patients were enrolled (mean age, 53.6 +- 14 years; 45% men; 67% black). There were 991 patients (31%) with an alternative diagnosis, 980 patients with a TIMI risk score of 0, and 828 with a TIMI score of 1. At low levels of TIMI risk (b3), adding in a Clinical impression of an alternative diagnosis did not reduce risk; at higher levels of TIMI risk, it did. The incidence of 30-day death, myocardial infarction, or revascularization for patients with a clinical impression of an alternative diagnosis and a TIMI score of 0 was 2.9% (95% confidence interval, 1.6%-5.0%).

Conclusions: The TIMI risk score stratifies patients with and without an alternative diagnosis. Unfortunately, patients with both a low TIMI risk score and a clinical impression of an alternative noncardiac diagnosis still have a risk of 30-day adverse events that is not low enough to allow safe discharge from the ED.

(C) 2009

* Corresponding author. Department of Emergency Medicine, Ground Ravdin, Hospital of the University of Pennsylvania, 3400 Spruce Street, Philadelphia, PA 19104. Tel.: +1 215 662 2767; fax: +1 215 662 3953.

E-mail address: [email protected] (A.M. Chang).

Introduction

Background

Effective risk stratification for patients who present to the emergency department (ED) with a complaint of chest pain is

0735-6757/$ - see front matter (C) 2009 doi:10.1016/j.ajem.2008.01.028

essential because the events may represent a life-threatening emergency or a nonurgent condition that only requires outpatient follow-up. Numerous studies have evaluated the effectiveness of risk stratification tools to identify chest pain patients safe for ED discharge [1-9]. Although the Thrombo- lysis in Myocardial Infarction (TIMI) risk score was derived from trials of patients with unstable angina or non-ST- segment elevation myocardial infarction, it also risk stratifies broad-based ED chest pain patient populations [10,11].

The ED chest pain patients with low TIMI scores (range, 0-1) still have a 3% to 5% risk of 30-day adverse cardiovascular events, and therefore, use of the TIMI risk score alone is not advocated to determine patient disposition [10,11]. A clinical impression of a clear-cut alternative noncardiac diagnosis identifies patients at reduced risk of both in-hospital and 30-day adverse events, but similarly does not identify a group of patients at such low risk that discharge can be advocated [12].

The goal of our investigation was to assess whether patients with a low TIMI risk score and a clinical impression of a clear-cut alternative noncardiac diagnosis would have a less than 1% risk for 30-day adverse cardiovascular events. We hypothesized that the presence of a low-risk TIMI risk score designed to predict risk for unstable coronary disease along with a clinical impression that the presenting complaint had a specific noncardiac etiology might identify a group of patients at such low risk for cardiovascular events that they could be safely released from the ED.

Methods

Study design

We performed a prospective observational cohort study to determine the 30-day event rates in ED chest pain patients who were initially considered to have a potential acute coronary syndrome (ACS) [13]. Patients were stratified by their TIMI risk scores and whether they had a clinical impression of a clear-cut alternative noncardiac diagnosis at the end of their ED evaluation. The University of Pennsylvania Committee on Research Involving Human Subjects approved the protocol.

Setting

The study was conducted at the Hospital of the University of Pennsylvania, an adult, urban, academic ED with an annual census of 55000 patients.

Selection of participants

The study population consisted of patients 30 years or older, who presented to the ED with a chief complaint of

chest pain and received an electrocardiogram to evaluate the possibility of an ACS. The decision to order an electro- cardiogram was made by the health care team. Patients were enrolled by trained research assistants 17 hours a day between 7:00 AM and midnight. Coverage during these times captures 85% to 95% of eligible patients at our institution.

Patients were eligible whether or not they were admitted to the hospital. We excluded patients who were non-English speaking or unable or unwilling to provide informed consent. More than 95% of the patients presenting to our ED are English speaking. Twelve percent of patients refused to provide informed consent.

Data collection and processing

Data were collected in accordance with Standardized reporting guidelines [14]. Treating physicians recorded duration of chest pain, chest pain onset, cardiac risk factors, TIMI risk score, prior Stress testing [15] and/or cardiac catheterization, and initial electrocardiogram interpretation. In addition, trained research assistants obtained demographic information including age, sex, and race [16]. At the conclusion of the evaluation, although the patient was still in the ED, the treating physician was asked whether the patient had a clear-cut alternative noncardiac diagnosis. If they responded yes, they were asked to categorize the diagnosis as gastroesophageal reflux disease, musculoskele- tal, or other, and to specify the diagnosis.

Admitted patients were followed daily during hospitaliza- tion for death and cardiac complications. Evaluation of hospital course was made through daily communication between investigators and the health care team, with postdischarge medical review when necessary. Patients were followed for 30 days to determine whether they were admitted to the hospital. Telephone interview and medical record review were used to assess outcomes.

Main outcomes

The main outcome was 30-day cardiovascular events, which was a composite of death, acute myocardial infarction, and revascularization (either percutaneous coronary inter- vention or Coronary artery bypass grafting) either during hospitalization or the subsequent 30 days. In addition, we prespecified an acceptable safety threshold for consideration of ED discharge of less than 1% rate of 30-day adverse events, in accordance with previous studies.

Data analysis

Data are presented with percent frequency of occurrence or means/medians, with 95% confidence intervals (CIs) provided for the main outcomes. Comparisons between patients who were and were not felt to have an alternative diagnosis were performed with Student t test or ?2 or Fisher

Age (y)	51.2 +- 15.0		54.7 +- 14.1		b.01
Sex (female)	553	59	1124	54	.01
Race					.16
Black	645	69	1370	66
White	258	28	645	31
Other	34	4	69	3
pain duration (IQR)	300 min (20-2880)		120 min (15-550)		b.01
Pain onset before ED arrival (IQR)	720 min (120-2880)		240 min (60-960)		b.01
Cardiac risk factors
Hypertension	455	48	1316	63	b.01
Diabetes	164	17	474	23	b.01
Elevated cholesterol level	185	20	671	32	b.01
Family history of CAD	93	10	341	16	b.01
Tobacco use	359	38	891	43	.02
TIMI risk factors
Age N65 y	178	19	481	23	.01
Known coronary disease	100	11	490	23	b.01
N3 cardiac risks factors	188	20	647	31	b.01
Aspirin use in past wk	202	22	800	38	b.01
2 or more episodes of angina	141	15	773	37	b.01
ST-segment deviation	23	2	118	6	b.01
Elevated cardiac markers	27	3	115	6	b.01
Prior stress testing	177	19	725	35	b.01
Abnormal findings	22	12	146	20	.04
Prior catheterization	123	13	536	26	b.01
Abnormal findings	45	36	279	52	b.01
Initial ECG interpretation					b.01
Normal/nonspecific	806	86	1593	76
Abnormal/nondiagnostic	72	8	188	9
Ischemia/MI	59	6	308	15
CAD indicates coronary artery disease; ECG, electrocardiogram; IQR, interquartile range; MI, myocardial infarction.

exact tests, as appropriate. Relative risks and 95% CIs are also presented. Data were analyzed using SAS statistical software (Version 9.1, SAS Institute, Cary, NC).

Table 1 Characteristics of patients stratified by presence or absence of alternative diagnosis

Patient characteristics Alternative diagnosis (n = 991)

No alternative diagnosis (n = 2178)

n

P

n

%

%

Results

Characteristics of study subjects

There were 3169 patients enrolled in the study (Table 1). Mean (+-SD) age of the population was 53.6 +- 14.5 years, 45% were men, and 67% were black. There were 991 patients (31%) with an alternative diagnosis, most commonly diagnosed with musculoskeletal disease (29%) or gastroesophageal reflux disease (15%). The remaining 2178 patients did not have a clear-cut alternative noncardiac diagnosis.

During hospitalization, there were 23 patients (b1%) who died, 115 (4%) sustained a myocardial infarction, and 114 (4%) received revascularization (of whom 18 [b1%] received

coronary artery bypass grafting) during the initial hospita- lization. Thirty-day follow-up was completed in 3028 patients (96%) and was not different between groups. By the time of the 30-day follow-up, a total of 48 patients (1.6%) died, and 15 patients (0.5%) received revascularization after hospital discharge.

Compared with patients with a higher TIMI risk score, patients with a low TIMI risk score (0 or 1) were less likely to meet the triple Composite outcome of death, myocardial infarction, or revascularization during hospitalization (2.6% vs 12.5%); absolute difference, -9.86% (95% CI, -7.95% to

-11.95%); risk ratio, 0.21 (95% CI, 0.15-0.29) or within

30 days (3.4% vs 14.2%); absolute difference, -10.75% (95% CI, -8.7% to -13.0%); risk ratio, 0.21 (95% CI, 0.18-

0.31). The incidence of 30-day death or cardiovascular events did not meet our prespecified threshold for safety (b1%), even in the low-risk group.

Likewise, patients with an alternative diagnosis were also less likely to meet the triple composite outcome of death, myocardial infarction, or revascularization during

TIMI score	Alternative diagnosis	No alternative diagnosis
0	13/443 (2.9)	11/537 (2.0)
1	14/268 (5.5)	24/560 (4.3)
2 3	13/133 (9.8) 8/63 (12.7)	32/447 (7.2) 39/285 (13.7)

hospitalization (3.7% vs 7.9%); absolute difference, -4.1%

Table 2 Relationship between TIMI score and presence or

absence of alternative diagnosis with respect to triple composite outcome

4

5-7

3/24 (12.5)

1/8 (12.5)

48/199 (24.1)

23/50 (46.0)

Values are expressed as n (%).

(95% CI, -2.4% to -5.8%); risk ratio, 0.47 (95% CI, 0.33-

0.68) or within 30 days (5.5% vs 8.8%); absolute difference,

-3.2% (95% CI, -1.2% to -5.1%); risk ratio, 0.63 (95% CI,

0.47-0.85). This incidence of 30-day death or cardiovascular events did not meet our prespecified threshold for safety (b1%), even for patients with an alternative diagnosis.

With respect to our main outcome, comparison of patients with a clinical impression of an alternative diagnosis across TIMI scores showed that the risk of 30-day adverse events was not altered by clinical impression of an alternative diagnosis in the patients with low TIMI risk scores, but at higher TIMI risk scores, a clinical impression of an alternative diagnosis reduced risk (Tables 2 and 3). Patients with both a TIMI risk score of 0 and a clinical impression of an alternative diagnosis still had a risk of 30-day events of 2.9% (95% CI, 1.6%-5.0%). This incidence of 30-day death or cardiovascular events did not meet our prespecified threshold for safety (b1%).

Discussion

An effective tool for risk stratifying patients and identify- ing a patient population safe for ED discharge using a limited,

but effective, combination of patient characteristics is strongly desired because emergency physicians are expected to provide a rapid error-free diagnosis and disposition.

The TIMI risk score [6] uses 7 items that are typically available in the ED: older than 65 years, documented prior coronary artery stenosis greater than 50%, 3 or more conventional cardiac risk factors, aspirin use within preced- ing 7 days, 2 or more anginal events in the past 24 hours, ST- segment elevation or depression greater than 1 mm, and elevated cardiac biomarkers [17]. Antman et al [6] derived and validated the TIMI risk score as a tool for predicting adverse cardiac outcome (death, [re]infarction, or revascu- larization) within 14 days in a patient population with unstable angina or non-STEMI. Pollack et al [11] and Chase et al [10] have independently applied the TIMI risk score to the unselected cohort of chest pain patients that present to the ED. They found that TIMI risk score could be used to risk stratify this broad patient population for both immediate (in- hospital) and 30-day Cardiovascular complications. The overall risk of 30-day adverse events ranged from 2.1% for patients with a TIMI score of 0 or 1, to greater than 45% for patients with a TIMI score of 6 or 7 [6,11]. Similarly, Chase et al [10] found that TIMI risk score was related to 30-day outcomes. However, even patients with a TIMI risk score of 0 had a 1.7% incidence of adverse outcomes. Both studies highlight the fact that TIMI risk score cannot be used in isolation to guide disposition.

Multiple clinical algorithms have been proposed for risk stratification of ED patients with potential ACSs, but none have been effective at identifying patients safe for discharge [8,9,11,18-28]. Hollander et al [12] hypothesized that the incorporation of a strong clinical impression of a clear-cut alternative noncardiac diagnosis, when ACSs was the only serious diagnostic concern, might identify a patient popula- tion safe for discharge from the ED. Patients with a clinical impression of a clear-cut alternative noncardiac diagnosis were less likely to experience a 30-day adverse event (death, acute myocardial infarction, or revascularization), but these

Table 3 Relationship between TIMI score and presence or absence of alternative diagnosis with respect to individual outcomes
TIMI score	Alternative diagnosis	In-hospital Death	MI	Revasc		30-d outcomes Death	MI	Revasc
0	Yes (n = 443)	8 (1.8)	1 (0.2)	1 (0.2)		0	12 (2.7)	0
	No (n = 537)	0	6 (1.1)	4 (0.7)		0	2 (0.4)	0
1	Yes (n = 268)	3 (1.1)	3 (1.1)	2 (0.8)		0	9 (3.4)	1 (0.3)
	No (n = 560)	0	16 (2.9)	12 (2.1)		0	1 (0.2)	2 (0.4)
2	Yes (n = 133)	3 (2.2)	6 (6.0)	3 (2.2)		0	5 (3.8)	1 (0.8)
	No (n = 447)	0	18 (4.0)	13 (2.9)		2 (0.4)	4 (0.9)	2 (0.4)
3	Yes (n = 63)	3 (4.8)	4 (6.3)	1 (1.6)		0	5 (3.8)	1 (0.8)
	No (n = 285)	2 (0.7)	18 (6.3)	25 (8.8)		1 (0.3)	4 (1.4)	3 (1.1)
4	Yes (n = 24)	0	2 (8.3)	0		0	2 (8.3)	0
	No (n = 199)	4 (2.0)	23 (11.6)	35 (17.6)		3 (1.6)	5 (2.5)	3 (1.5)
5-7	Yes (n = 8)	0	0	0		0	0	1 (12.5)
	No (n = 61)	0	26 (42.6)	18 (29.5)		2 (3.3)	0	1 (1.6)
Revasc indicates revascularization.

patients still had a 4% risk of an adverse event. The risk of adverse events using clear-cut alternative noncardiac diag- nosis alone was not reduced to an acceptable level to identify patients safe for ED discharge.

We hypothesized that because both of these approaches were able to successfully risk stratify patients, perhaps the combination of a low TIMI risk score and an alternative diagnosis could take it 1 step further and identify the cohort of patients with a less than 1% risk of 30-day adverse events. We found that for higher TIMI risk scores, a clinical impression of a noncardiac diagnosis was associated with a reduced risk, but that the combination could still not identify low-risk patients at less than 1% risk.

The current standard of care requires admission to the hospital or observation unit for at least 8 to 12 hours for most of the patients with potential ACS [29]; however, recent small studies suggest that computerized tomographic coronary angiography may be able to successfully identify patients at such low risk of adverse events that they can be safely discharged from the ED without a traditional rule- out protocol [30,31]. There are large ongoing clinical studies that will clarify whether this strategy is safe and cost effective.

There are several possible limitations to our study. Our study population was an urban ED chest pain patient population, with more black and female patients than the groups from which the TIMI score was derived. Therefore, it may be difficult to generalize our results to all patient populations. Furthermore, our study population represents only those patients who presented to the ED with symptoms suggestive of an ACS. These results may not be generalized to patients evaluated electively in an office-based setting. Our study goal was related to improving identification of low-risk patients, where the combination was not effective. We found that there was some use of the combination toward reducing risk in the high-risk group, but our sample size was relatively small in this subset.

In conclusion, we found that although both the TIMI risk score and the clinical impression of a clear-cut alternative noncardiac diagnosis can risk stratify patients with potential ACS in the ED, even a combination of both cannot identify patients at less than 1% risk of 30-day events.

References

1. Jaffery Z, Hudson MP, Jacobsen G, Nowak R, McCord J. Modified Thrombolysis In Myocardial Infarction risk score to risk stratify patients in the emergency department with possible acute coronary syndromes. J Thromb Thrombolysis 2007;24(2):137-44.
2. Yan AT, Yan RT, Tan M, et al. Risk scores for risk stratification in acute coronary syndromes: useful but simpler is not necessarily better. Eur Heart J 2007;28(9):1072-8.
3. Calvin JE, Klein LW, VandenBerg BJ, et al. Risk stratification in unstable angina: prospective validation of the Braunwald classifica- tion. JAMA 1995;273:136-41.
4. Jacobs Jr DR, Kroenke C, Crow R, et al. PREDICT: a simple risk score for clinical severity and long-term prognosis after hospitalization for acute myocardial infarction or unstable angina: the Minnesota heart survey. Circulation 1999;100:599-607.
5. Boersma E, Pieper KS, Steyerberg EW, et al. Predictors of outcome in patients with acute coronary syndromes without persistent ST-segment elevation. Results from an international trial of 9461 patients. Circulation 2000;101:2557-67.
6. Antman EM, Cohen M, Bernink PJ, 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.
7. Granger CB, Goldberg RJ, Dabbous O, et al. Predictors of hospital mortality in the global registry of acute coronary events. Arch Intern Med 2003;163:2345-53.
8. Baxt WG, Shofer FS, Sites FD, Hollander JE. A neural computational aid to the diagnosis of acute myocardial infarction. Ann Emerg Med 2002;39(4):366-73.
9. Baxt WG, Shofer FS, Sites FD, Hollander JE. A neural network aid for early diagnosis of cardiac ischemia in patients presenting to the emergency department with chest pain. Ann Emerg Med 2002;40(6): 575-83.
10. Chase M, Robey JL, Zogby KE, Sease KL, Shofer FS, Hollander JE. Prospective validation of the Thrombolysis in Myocardial Infarction risk score in the emergency department chest pain population. Ann Emerg Med 2006;48(3):252-9.
11. Pollack CV Jr, Sites FD, Shofer FS, Sease KL, Hollander JE. Application of the TIMI risk score for unstable angina and non-ST elevation acute coronary syndrome to an unselected emergency department chest pain population. 2006;13(1):13-8.
12. Hollander JE, Robey JL, Chase MR, Brown AM, Zogby KE, Shofer FS. Relationship between a clear-cut alternative noncardiac diagnosis and 30-day outcome in emergency department patients with chest pain. Acad Emerg Med 2007;14(3):210-5.
13. Miller CD, Lindsell CJ, Khandelwal S, et al. Is the initial diagnostic impression of “noncardiac chest pain” adequate to exclude cardiac disease? Ann Emerg Med 2004;44:565-74.
14. Hollander JE, Blomkalns AL, Brogan X, 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.
15. Nerenberg RH, Shofer FS, Robey JL, Brown AM, Hollander JE. Impact of a negative prior stress test on emergency physician Disposition decision in ED patients with chest pain syndromes. Am J Emerg Med 2007;25(1):39-44.
16. Hollander JE, Valantine SM, Brogan GX. The Academic Associate Program: integrating clinical emergency medicine research with undergraduate education. Acad Emerg Med 1997;4:225-30.
17. Manenti ER, Bodanese LC, Camey SA, Polanczyk CA. Prognostic value of serum biomarkers in association with TIMI risk score for acute coronary syndromes. Clin Cardiol 2006;29(9):405-10.
18. Goldman L, Cook EF, Brand DA, et al. A computer protocol to predict myocardial infarction in emergency department patients with chest pain. N Engl J Med 1988;318:797-803.
19. Goldman L, Cook EF, Johnson PA, Brand DA, Rouan GW, Lee TH. Prediction of the need for intensive care in patients who come to emergency departments with acute chest pain. N Engl J Med 1996;334: 1498-504.
20. Pozen MW, D’Agostino RB, Selker HP, Sytkowski PA, Hood WB. A predictive instrument to improve coronary care unit admission practices in acute ischemic heart disease. N Engl J Med 1984;310: 1273-8.
21. Hollander JE. The continuing search to identify the very-Low-risk chest pain patient. Acad Emerg Med 1999;6:979-81.
22. Fineberg HV, Scadden D, Goldman L. Care of patients with a low probability of acute myocardial infarction. cost effectiveness of alternative to coronary care admission. N Engl J Med 1984;310: 1301-7.
23. Pope JH, Aufderheide TP, Ruthazer R, et al. Missed diagnosis of Acute cardiac ischemia in the emergency department. N Engl J Med 2000; 342:1163-70.
24. Goldman L, Weinberg M, Weisberg M, 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.
25. Tatum JL, Jesse RL, Kontos MC, et al. Comprehensive strategy for the evaluation and triage of the chest pain patient. Ann Emerg Med 1997; 29:116-25.
26. Selker HP, Beshansky JR, Griffith JL, 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. Ann Intern Med 1998;129:845-55.
27. Selker HP, Zalenski RJ, Antman EM, et al. An evaluation of technologies for identifying acute cardiac ischemia in the emergency department: a report from the National Heart Attack Alert Program Working Group. Ann Emerg Med 1997;29:13-81.
28. Christenson J, Innes G, McKnight D, et al. A clinical prediction rule for early discharge of patients with chest pain. Ann Emerg Med 2006;47:1-10.
29. Anderson JL, Adams CD, Antman EM, et al. ACC/AHA 2007 guidelines for the management of patients with unstable angina/non- ST-elevation myocardial infarction. A report of the American College of Cardiology/American Heart Association Task Force on practice guidelines (Writing Committee to Revise the 2002 Guidelines for the Management of Patients With Unstable Angina/non-ST-elevation myocardial infarction). J Am Coll Cardiol 2007;50:1-157.
30. Hollander JE, Litt HI, Chase M, Brown AM, Kim W, Baxt WG. Computed tomography coronary angiography for rapid disposition of low risk emergency department patients with chest pain syndromes. Acad Emerg Med 2007;14:112-6.
31. Goldstein JA, Gallagher MJ, O’Neill WW, Ross MA, ONeill BJ, Raff GL. A randomized controlled trial of multislice coronary computed tomography for evaluation of acute chest pain. J Am Coll Cardiol 2007;49:863-71.