American Journal of Emergency Medicine (2012) 30, 1706-1711

Original Contribution

Utilization of coronary computed tomography angiography for exclusion of coronary artery disease in ED patients with low- to intermediate-risk chest pain: a 1-year experience^?

Adam J. Singer MD ^a,?, Anna Domingo BS ^a, Henry C. Thode Jr. PhD ^a,

Melissa Daubert MD ^b, Alan F. Vainrib MD ^b, Summer Ferraro ^b, Amee Minton ^b,

Annie Poon BA ^a, Mark C. Henry MD ^a, Michael Poon MD ^b

^aDepartment of Emergency Medicine, Stony Brook University, Stony Brook, NY

^bDepartment of Radiology, Stony Brook University, Stony Brook, NY

Received 19 June 2011; revised 18 January 2012; accepted 22 January 2012

Abstract

Objective: We describe our preliminary experience with coronary computed tomography angiography in emergency department (ED) patients with low- to intermediate-risk chest pain.

Methods: A convenience cohort of patients with low- to intermediate-risk acute chest pain presenting to a suburban ED in 2009 were prospectively enrolled if the attending physician ordered a CCTA for possible coronary artery disease. Demographic and clinician data were entered into structured data collection sheets required before any imaging. The results of CCTA were classified as normal, nonobstructive (1%-50% stenosis), and obstructive (N50% stenosis). Outcomes included hospital admission and death within a 6-month follow-up period.

Results: In 2009, 507 patients with ED chest pain had a CCTA while in the ED. The median (interquartile range) age was 54 (47-62) years; 51.5% were female. Thrombolysis in Myocardial Infarction risk scores were 0 (42.6%), 1 (42.2%), 2 (11.8%), 3 (2.4%), and 4 (1.0%). The results of CCTA were normal (n = 363), nonobstructive (n = 123), and obstructive (n = 21). Admission rates by CCTA results were obstructive (90.5%), nonobstructive (4.9%), and normal (3.0%). None of the patients with normal or nonobstructive CCTA died within the 6-month follow-up period (0%; 95% confidence interval, 0-0.9%).

Conclusions: Many ED patients with low- to intermediate-risk chest pain have a normal or nonobstructive CCTA and may be safely discharged from the ED without any associated mortality within the following 6 months.

(C) 2012

Introduction

^? Presented in part at the American College of Physicians Research Forum, September 2010, Las Vegas, NV.

* Corresponding author. Tel.: +1 631 444 7857; fax: +1 631 444 7919.

E-mail address: [email protected] (A.J. Singer).

Each year, there are nearly 6 million patients presenting to US emergency departments (EDs) with chest pain [1]. Although most such patients have noncardiac chest pain, up to a third will be diagnosed as having chest pain of cardiac ischemic nature [2]. Prior studies suggest that 2% to 6% of

0735-6757/$ - see front matter (C) 2012 http://dx.doi.org/10.1016/j.ajem.2012.01.025

ED patients with acute coronary syndromes are inappropri- ately misdiagnosed and discharged with an increased likelihood of adverse events and subsequent litigation [3- 5]. These concerns have led many emergency physicians to adopt a very conservative approach in which most patients with nontraumatic chest pain are further evaluated and/or admitted [6]. Because it is currently impossible to identify all patients at Very low risk of adverse cardiac events that could be rapidly discharged from the ED without further testing, many low-risk patients undergo prolonged observation and advanced cardiac testing and imaging in the ED or inpatient settings. Until recently, many such low-risk patients with chest pain underwent serial biomarker testing lasting anywhere from 6 to 12 hours with or without additional cardiac imaging in the hospital or shortly after discharge. The holy grail of chest pain has remained a rapid, highly accurate, noninvasive, cost-effective test that allows exclusion of acute coronary syndromes and early patient discharge from the ED [7].

Multidetector coronary computed tomography angiogra- phy (CCTA) has recently been added to an expanding armamentarium of noninvasive cardiac imaging modalities [7]. In addition to its ability to visualize coronary artery luminal patency, it can also visualize nonobstructive plaque and calcifications. It is quick and relatively simple and can be completed within 10 minutes, allowing visualization of the entire coronary artery tree while also supplying information on non-coronary artery pathologies. Coronary computed tomography angiography is also recommended as an alternative to conventional Stress testing by the American College of Cardiology/American Heart Association in the evaluation of acute chest pain [8]. A recent meta-analysis of more than 2000 patients has demonstrated that CCTA accurately diagnoses significant coronary artery stenosis (N50% luminal narrowing), with a sensitivity of 96% to 98%, a specificity of 85% to 89%, a negative predictive value of 94% to 97%, and a positive predictive value of 93% [9]. An even larger meta-analysis involving 9592 patients pooled from 18 studies with suspected coronary artery disease (CAD) found a sensitivity of 99% and specificity of 41%. In addition, it showed a low rate of Major adverse cardiac events after 20-month follow-up in those with less than 50% coronary luminal stenosis (nonobstructive). The pooled annualized event rate for obstructive vs normal CCTA was 8.8% vs 0.17% per year for major adverse cardiac events (P b .05) and 3.2% vs 0.15% for death or myocardial infarction (MI; P b.05) [10].

Table 1 Characteristics of patient sample

Several ED-based single-center and multicenter studies have shown that the risk of major adverse cardiac events in patients with low-to-intermediate risk chest pain with a negative CCTA is very low [11-13]. In these studies, CCTA was generally available only during traditional business hours on weekdays, and none has reported on the impact of CCTA on admission rates for low- to intermediate-risk patients with chest pain. Recent changes in reimbursement rates have discouraged hospitals from admitting ED patients

with Low-risk chest pain and have encouraged using observation in the ED to exclude significant CAD. With increasing ED crowding, efforts to improve the process of care and reduce ED lengths of stay without negatively impacting patient outcomes are sorely needed. With this in mind, our institution has adopted CCTA in an effort to reduce admission rates while shortening length of stay when evaluating ED patients with low- to intermediate-risk chest pain.

The goals of the current study were to describe our preliminary experience with CCTA in a cohort of ED patients with low- to intermediate-risk chest pain when used on a daily basis and to determine its impact on the rate of admissions of such patients. Prior studies have been limited to traditional business hours and have not reported the impact on admission rates. We also determined the mortality rate in patients with normal and nonobstructive CCTA during a 6-month period subsequent to coronary imaging.

Methods

Study design

A prospective Observational study design was used to enroll patients in the study. The study was approved by the institutional review board. Because the study was

Characteristics Cases in the sample (n = 507), unless otherwise specified

Female, no. (%) 261 (51.5) Age (y), median (IQR) 54 (47-62) Hypertension, no. (%) 190 (37.5) Diabetes, no. (%) 36 (7.1) Hyperlipidemia, no. (%) 134 (26.4) Ever smoked, no. (%) 167 (32.9) Family history of CAD, no. (%) 213 (42.3) a History of CAD, no. (%) 12 (2.4) b Prior use of aspirin, no. (%) 79 (15.6) b Multiple episodes of chest pain, no. (%) 175 (34.5) TIMI risk score 0 216 (42.6%) 1 214 (42.2%) 2 60 (11.8%) 3 12 (2.4%) 4 5 (1.0%) Calcium score, median (IQR) 11 (0-139) c

a Excludes 4 patients with missing/unknown family history and 4 patients who were adopted and did not know family history. b Excludes 1 patient with unknown prior aspirin use and history of CAD. c Excludes 8 patients without a calcium score.

implemented as part of a continuous Quality improvement project, waiver from written informed consent was given by the Committee on Research Involving Human Subjects.

Setting

The study was conducted at a tertiary care university- based subUrban academic medical center with an annual ED census of approximately 90 000. Our ED introduced 64-slice multidetector CCTA in January 2009. There was no run-in period before enrolling the first study patient.

Currently, our ED does not have a dedicated chest pain unit (scheduled to open February 2012). Thus, patients with chest pain awaiting a disposition remain in the general ED, usually in our critical care area. Before introducing CCTA, patients were usually admitted to the hospital for serial biomarker testing, a subsequent provocative testing if troponins remained normal after 8 to 12 hours. After introducing CCTA, patients in whom

a CCTA was ordered were placed on Observation status but remained in the ED until results of the CCTA were reported by a board-certified radiologist or cardiologist specializing in CCTA.

Subjects

Patients were eligible for the study if they had chest pain of suspected cardiac ischemic origin and if the treating attending ED physician ordered a CCTA. At our hospital, CCTAs are obtained in patients with an initial negative troponin I level (<=0.04 pg/mL) and a normal or nondiag- nostic 12-lead electrocardiogram. Patients with ST-segment elevation, ST depressions greater than or equal to 1 mm, positive cardiac Troponin I levels, and known CAD (including a history of MI, percutaneous coronary interven- tion [PCI], or coronary artery bypass graft [CABG]) are excluded from CCTA. Patients with a thrombolysis in MI (TIMI) score greater than 4 are also excluded from CCTA.

Cardiac CTA Subject Flow Sheet

579 Patients Total:

72 Excluded

o48 Non-Ed CTAs

o9 Refused Consent

o15 High Risk

Low to Intermediate Risk Patients

507

Normal CCTA 363

Non-Obstructive CCTA 123

Obstructive CCTA 21

Admitted 11

Discharged 352

Admitted 6

Discharged 117

Admitted 19

Discharged 2

Catheterization 0

Catheterization 1

No Catheterization 5

Catheterization 11

No Catheterization 8

Obstructed with PCI

Obstructed 9

Non-Obstructed 2

PCI (7)

CABG (1)

No Intervention (1)

CCTA: Computed coronary tomography angiography PCI: Percutaneous coronary intervention

CABG: Coronary artery bypass graft

Fig. 1 Study patient ftow.

Characteristic	Normal CCTA (n = 363)	Nonobstructive CCTA (n = 123)	Obstructive CCTA (n = 21)	P
Age (y), median (IQR)	45 (39-51)	53 (46-63)	59 (51-67)	b.001
% Female	57.6	36.6	33.3	b.001
% Hypertension	30.6	52.8	66.7	b.001
% Diabetes	6.3	8.1	14.3	.34
% Hyperlipidemia	21.2	35.0	66.7	b.001
% Ever smoked	29.8	38.2	60.0	.003
% Family history CAD a	41.0	44.3	52.4	.52
% History of CAD b	0.6	8.1	0.0	b.001
% Prior aspirin use b	13.0	21.1	28.6	.02
% Multiple chest pain episodes	36.4	29.3	33.3	.36
TIMI risk score, median (IQR)	1 (0-1)	1 (0-2)	1 (1-3)	b.001
Mean calcium score c	0 (0-0)	53 (22-145)	399 (216-1025)	b.001
a Excludes 4 patients with missing/unknown family history and 4 patients who were adopted and did not know family history. b Excludes 1 patient with unknown prior aspirin use and history of CAD. c Excludes 8 patients without a calcium score.

Patients in whom standard coronary artery angiography was planned were also excluded.

Table 2 Characteristics of patients by CCTA result

Measurements

Standardized data collection was performed prospectively using forms that included patient demographics, clinical history, and presence of traditional cardiac risk factors [14,15]. The CCTA technicians were instructed not to perform the test unless a data collection form was completed by the ordering physician. For all patients, telephone follow- up or search of the social security death index was performed to determine whether the patient had died during the follow- up period.

Outcomes

The primary outcomes were the Hospital admission rates and the mortality rate during the 6-month study period subsequent to cardiac imaging.

Data analysis

Binary data are presented as the percent frequency of occurrence and compared with a ?2 or Fisher exact test as appropriate. Continuous data are presented as medians and interquartile ranges (IQRs) and compared using Kruskal- Wallis tests. Data analysis was performed using PASW Statistics 18 software (SPSS Inc, Chicago, IL).

Results

During the 2009 calendar year, 507 ED patients with low- to intermediate-risk chest pain had a CCTA while in the ED.

Their median (IQR) age was 54 (47-62) years; 51.5% were female (Table 1). Coronary risk factors included family history of CAD (42.3%), hypertension (37.5%), hyperlipid- emia (26.4%), smoking (32.9%), and diabetes (7.1%). Seven percent of all these study patients were admitted to the hospital. A distribution of TIMI risk score is presented in Table 1. Most had a score of 0 or 1. No patients had a TIMI risk score greater than 4.

Of all patients undergoing CCTA in 2009, 363 (71.6%) had Normal coronary arteries, 123 (24.3%) had nonobstruc- tive CAD, and 21 (4.1%) had obstructive CAD (Fig. 1). Age was associated with the degree of coronary artery obstruction (normal [median age, 45 years], nonobstructive [53 years], and obstructive [59 years]; Pb.001). Sex was also associated with CCTA result (P b .001), with the majority of patients with normal CCTA being female (58%) and the minority of patients with nonobstructive and obstructive CCTAs being female (36.6% and 33.3%, respectively). Hypertension, hyperlipidemia, smoking, history of CAD, and prior aspirin use were all significantly related to CCTA results. Diabetes, family history of CAD, and multiple episodes of chest pain were not associated with CCTA results (Table 2).

Of the 486 patients with normal or nonobstructive CCTAs, 469 (96.7%) were discharged home from the ED after the interpretation of their CCTA compared with 9.5% for those with an obstructive CCTA (Table 3). Most

Table 3 Patient disposition by CCTA result

CCTA result	No. of patients admitted	% Admitted
Normal (n = 363)	11	3.0
Nonobstructive (n = 123)	6	4.9
Obstructive (n = 21)	19	90.5
All (N = 507)	36	7.1

PCI	36.8%	16.7%	0%	.05
CABG	5.3%	0%	0%	.62
MI	0%	0%	0%	-

patients with obstructive CAD who were discharged had a cardiology consult resulting in an outpatient stress test, and a few such patients signed out against medical advise. For admitted patients, there was no significant difference in the rate of CABG procedures or of the incidence of MI by CCTA result (Table 4). None of the admitted patients died while in the hospital.

Table 4 Procedures/outcomes of patients admitted at initial hospitalization by CCTA result

Obstructive Nonobstructive (n = 19) (n = 6)

Normal (n = 11)

P

Follow-up

Six-month telephone follow-up was completed in 346 patients (68.2%), and in the remainder of the patients, a search of the social security death registry was performed. There were no deaths of any patients (overall mortality rate, 0%; 95% confidence interval, 0-0.9%). Of the 346 patients who were contacted, none with normal or nonobstructive CCTA had a subsequent PCI or CABG.

Discussion

As health care reform evolves, there have been and will continue to be changes in the rates of reimbursement for patients with low-risk chest pain who are admitted to the hospital. At many institutions, hospitals are no longer reimbursed for such short admissions, and there has been a shift toward completing the evaluation of such patients during an ED observation period. With this in mind, our institution has adopted CCTA to allow rapid and safe exclusion of obstructive CAD, significantly reducing the need for admission. Our results show that most ED patients with low- to intermediate-risk chest pain can be rapidly “ruled out” and discharged home from the ED when CCTA is used to exclude obstructive CAD. This approach is very safe because none of the patients with normal or nonobstructive CAD who were discharged from the ED died in the subsequent 6 months. Use of CCTA in this patient population also resulted in a very low rate of admissions to the hospital for ED patients with chest pain.

Our results are in agreement with several other single- center and multicenter studies demonstrating the accuracy and safety of CCTA in excluding significant CAD in ED patients with low- to intermediate-risk chest pain. Hollander et al [11] observed 481 ED patients with a negative CCTA for 1 year after discharge. Among these patients, 11% were

rehospitalized and 11% underwent further diagnostic testing. There was only 1 death of unclear etiology, and there were no MIs or revascularization procedures during the study period. In a similar study from Korea, none of the 103 patients with normal CCTA developed any major adverse cardiac events during a 1-month follow-up period [12]. A randomized control study evaluating the use of CCTA vs standard of care enrolled 200 low- to intermediate-risk ED patients with acute chest pain. Of patients randomized to CCTA, two thirds had minimal or no CAD and were rapidly discharged from the ED. Compared with the standard of care, CCTA reduced the time to diagnosis from 15 to 3.4 hours, with no test complications or subsequent diagnosis of CAD or major adverse cardiac events over the next 6 months [13]. Rubinstein et al [16] demonstrated a low rate of major adverse cardiac events and favorable prognosis at 12-month follow-up in intermediate-risk patients with CCTA showing no or minimal coronary artery plaque. In the 2-year follow-up of the Rule Out Myocardial Infarction Using Computer- Assisted Tomography observational cohort, patients with stenosis greater than 50% by CCTA had a significantly higher rate of adverse events (15% vs 0.4%) and a significant increase in hazard ratio (hazard ratio, 50) compared with patients with less severe or no coronary artery stenosis [17]. In data presented from the Coronary computed tomographic angiography for Systematic Triage of Acute Chest Pain Patients to Treatment trial, 749 patients with acute chest pain were randomized to CCTA vs standard of care with nuclear myocardial perfusion imaging. This study showed that there was a shorter time to diagnosis (2.9 hours vs 6.2 hours, respectively), reduced hospital costs ($2138 vs $3458, respectively), and a decrease in radiation exposure (10.8 mSv vs 15 mSv, respectively) when a CCTA strategy is used [18].

Study limitations

A major limitation of this study was the relatively low follow-up rate. Thus, it is not possible to exclude major adverse cardiac events in patients that could not be contacted by telephone or who had not presented to our study institution. However, by reviewing death records, we were able to determine that none of the study patients died within the United States during the follow-up period. In addition, our center is the major cardiovascular center in our region. Thus, many patients with significant CAD are transferred to our center. Second, we did not determine whether all patients eligible for ED CCTA were included. In our experience, the use of CCTA by Emergency practitioners and cardiologists has increased greatly during the course of the study. As a result, we have recently added a 320 slice multidetector CT in our ED. Third, our study is limited to a single center with a highly qualified and dedicated staff with considerable experience with CCTA. As such, our results may not be generalizable to other settings or institutions.

In conclusion, in our institution, the use of CCTA for ED patients with chest pain of suspected cardiac ischemic origin is highly accurate and safe and may lead to a significant reduction in the number of admissions for such patients.

References

1. Nawar EW, Niska RW, Xu J. National Hospital Ambulatory Medical Care Survey: 2005 emergency department summary. Adv Data 2007(386):1-32.
2. Hollander JE. The continuing search to identify the very-low-risk chest pain patient. Acad Emerg Med 1999;6(10):979-81.
3. McCarthy BD, et al. missed diagnoses of acute myocardial infarction in the emergency department: results from a multicenter study. Ann Emerg Med 1993;22(3):579-82.
4. Pope JH, et al. Missed diagnoses of Acute cardiac ischemia in the emergency department. N Engl J Med 2000;342(16):1163-70.
5. Karcz A, et al. Massachusetts emergency medicine closed malpractice claims: 1988-1990. Ann Emerg Med 1993;22(3):553-9.
6. Katz DA, et al. Emergency physicians’ fear of malpractice in evaluating patients with possible acute cardiac ischemia. Ann Emerg Med 2005;46(6):525-33.
7. Hoffmann U, Bamberg F. Is Computed tomography coronary angiography the most accurate and effective noninvasive imaging tool to evaluate patients with acute chest pain in the emergency department? Response to Hoffman and Bamberg. Circ: Cardiovasc Imaging 2009;2(3):251-63.
8. Taylor AJ, et al. ACCF/SCCT/ACR/AHA/ASE/ASNC/NASCI/S- CAI/SCMR 2010 appropriate use criteria for cardiac computed tomo- graphy: a report of the American College of Cardiology Foundation Appropriate Use Criteria Task Force, the Society of Cardiovascular Computed Tomography, the American College of Radiology, the American Heart Association, the American Society of Echocardiog- raphy, the American Society of Nuclear Cardiology, the North American Society for Cardiovascular Imaging, the Society

for Cardiovascular Angiography and Interventions, and the Society for Cardiovascular Magnetic Resonance. J Am Coll Cardiol 2010; 56(22):1864-94.

1. Mowatt G, et al. 64-Slice computed tomography angiography in the diagnosis and assessment of coronary artery disease: systematic review and meta-analysis. Heart 2008;94(11):1386-93.
2. Hulten EA, et al. Prognostic value of cardiac computed tomography angiography a systematic review and meta-analysis. J Am Coll Cardiol 2010.
3. Hollander JE, et al. One-year outcomes following coronary comput- erized tomographic angiography for evaluation of emergency department patients with potential acute coronary syndrome. Acad Emerg Med 2009;16(8):693-8.
4. Kim J, et al. Efficacy and safety of the computed tomography coronary angiography based approach for patients with acute chest pain at an emergency department: one month clinical follow-up study. J Korean Med Sci 2010;25(3):466-71.
5. Goldstein JA, et al. A randomized controlled trial of multi-slice coronary computed tomography for evaluation of acute chest pain. J Am Coll Cardiol 2007;49(8):863-71.
6. Standardized reporting guidelines for studies evaluating risk stratifi- cation of emergency department patients with potential acute coronary syndromes. Ann Emerg Med 2004;44:589-98.
7. Hollander JE, Mayorga O, Sites FD, et al. Defining the outcomes of ED chest pain patient risk stratification studies: use of a standardized definition for unstable angina. Acad Emerg Med 2002;9:375.
8. Rubinstein R, et al. Impact of 64-slice cardiac computed tomographic angiography on clinical decision-making in emergency department patients with chest pain of possible myocardial ischemic origin. Am J Cardiol 2007;100(10):1522-6.
9. Hoffmann U, et al. Coronary computed tomography angiography for early triage of patients with acute chest pain: the ROMICAT (Rule Out Myocardial Infarction using Computer Assisted Tomography) trial. J Am Coll Cardiol 2009;53(18):1642-50.
10. Goldstein JA, Chinnaiyan KM, Abidov A, et al. The CT-STAT (coronary computed tomographic angiography for Systematic Triage of Acute Chest Pain Patients to Treatment) trial. J Am Coll Cardiol 2011;58:1414-22.