Coronary Computed Tomography Angiography Versus Traditional Care: Comparison of One-Year Outcomes and Resource Use

CARDIOLOGY/ORIGINAL RESEARCH

Coronary Computed Tomography Angiography Versus Traditional Care: Comparison of One-Year Outcomes and Resource Use Judd E. Hollander, MD*; Constantine Gatsonis, PhD; Erin M. Greco, MS; Bradley S. Snyder, MS; Anna Marie Chang, MD; Chadwick D. Miller, MD, MS; Harjit Singh, MD; Harold I. Litt, MD, PhD† *Corresponding Author. E-mail: [email protected].

Study objective: Three large, multicenter, randomized, clinical trials have shown that coronary computed tomography (CT) angiography allows efficient evaluation and safe discharge of patients with low- to intermediate-risk chest pain who present to the emergency department (ED). We report 1-year event rates and resource use from the American College of Radiology Imaging Network-Pennsylvania 4005 multicenter trial. Methods: Patients with low- to intermediate-risk chest pain and presenting to the ED were randomized in a 2:1 ratio to a coronary CT angiography care pathway or traditional care. Subjects were contacted by telephone at least 1 year after ED presentation. Medical record review was performed for all cardiac hospitalizations, procedures and diagnostic tests, and adverse cardiac events. Our main outcome was the composite of cardiac death and myocardial infarction within 1 year. The secondary outcome was resource use. Results: One thousand three hundred sixty-eight patients enrolled and 1,285 (94%) had direct participant or proxy contact at 1 year. All others had record review or death index search. From index presentation through 1 year, there was no difference between patients in the coronary CT angiography arm versus traditional care with respect to major adverse cardiac event (1.4% versus 1.1%; difference 0.3%; 95% CI –5.5% to 6.0%). From hospital discharge through 1 year, there was also no difference in ED revisits (36% versus 38%; difference –2.1%; 95% CI –7.9% to 3.7%), hospital admissions (16% versus 17%; difference –0.9%; 95% CI –6.7% to 4.9%), or subsequent cardiac testing (13% versus 13%; difference –0.4%; 95% CI –6.2% to 5.5%). One of 640 subjects with a negative coronary CT angiography result had a major adverse cardiac event within 1 year of presentation (0.16%; 95% CI 0.004% to 0.87%). Conclusion: A coronary CT angiography–based strategy for evaluation of patients with low- to intermediate-risk chest pain who present to the ED does not result in increased resource use during 1 year. A negative coronary CT angiography result is associated with a less than 1% major adverse cardiac event rate during the first year after testing. [Ann Emerg Med. 2015;-:1-9.] Please see page XX for the Editor’s Capsule Summary of this article. 0196-0644/$-see front matter Copyright © 2015 by the American College of Emergency Physicians. http://dx.doi.org/10.1016/j.annemergmed.2015.09.014

INTRODUCTION The optimal pathway for the evaluation of patients with low to intermediate risk who present to the emergency department (ED) with chest pain and possible acute coronary syndrome remains a topic of intense investigation and controversy. Three large multicenter trials (Coronary Computed Tomography for Systematic Triage of Acute Chest Pain Patients to Treatment,1 American College of Radiology Imaging Network-Pennsylvania [ACRIN PA] 4005,2 and Multicenter Study to Rule Out Myocardial Infarction by Cardiac Computed Tomography [ROMICAT] II3) have demonstrated, in aggregate, that †

All study participants are listed in the Appendix.

Volume

-,

no.

-

:

-

2015

a coronary computed tomography (CT) angiography–based pathway allows more efficient evaluation of these patients and safe discharge of those with a negative study result. Two of the three studies measured cost of care; both demonstrated lower ED costs, although the ROMICAT II trial showed higher inpatient costs in the coronary CT angiography arm but overall cost neutrality at 28 days. Although the rate of ED recidivism was unchanged at 30 days, there was concern over increased short-term resource use because patients in the coronary CT angiography arms of all 3 trials underwent more testing.4,5 However, this was primarily a result of the design of two of the studies, which mandated testing in the coronary CT angiography arm, Annals of Emergency Medicine 1

Hollander et al

One-Year Outcomes of ACRIN Trial

Editor’s Capsule Summary

What is already known on this topic Coronary computed tomography (CT) angiography can identify chest pain patients at low risk for 30-day adverse events. What question this study addressed The 1-year risk of major adverse cardiac events in patients who were enrolled in a coronary CT angiography clinical trial. What this study adds to our knowledge One of 640 patients (0.16%) with a negative coronary CT angiography result had a major adverse event at 1 year. Major adverse events and resource use at 1 year did not differ between the patients who received imaging in the emergency department and those who underwent traditional testing. How this is relevant to clinical practice In patients who have a negative CT angiography result, the incidence of major adverse cardiac events at 1 year is very low, and they may not benefit from repeated risk stratification. Given the low event rate, larger studies will be needed to determine the optimal strategy for the evaluation of low-risk chest pain patients.

when possible, but did not require testing in the traditional care arm. Although single-center trials have demonstrated longterm safety (
1 year) for patients after negative coronary CT angiography results,6,7 1-year event rates and resource use have not been reported for any multicenter randomized controlled trial, to our knowledge. We analyzed patient outcomes and resource use within 1 year for the ACRIN PA 4005 trial. MATERIALS AND METHODS Study Design and Selection of Participants The design of the ACRIN PA 4005 trial has been described previously.2 Briefly, 1,392 subjects presenting to EDs at 5 centers from July 2009 to November 2011 were randomized in a 2:1 ratio to a coronary CT angiography care pathway, in which the first evaluation was a coronary CT angiography, or a traditional care pathway, in which the subject’s health care provider selected which tests, if any, would be performed. Eligible subjects were aged at least 30 years, with a chief complaint consistent with potential acute 2 Annals of Emergency Medicine

coronary syndrome, an ECG not demonstrating acute ischemia, an initial Thrombolysis in Myocardial Infarction risk score8 of 0 to 2, and provider-determined need for admission or objective testing to exclude acute coronary syndrome. Objective testing included plans for a functional (stress) or anatomic (coronary CT angiography or catheterization) evaluation. Subjects were enrolled 7 AM to midnight, 7 days per week, and potential subjects presenting outside these hours were eligible to be enrolled the next morning if the manner of further testing had not already been decided on. Patients unable or unwilling to participate in follow-up were excluded from enrollment. The primary hypothesis of the trial was that patients without significant coronary artery disease on coronary CT angiography have a less than 1% rate of 30-day cardiac death or myocardial infarction. This report focuses on major secondary aims, including comparison of 1-year rates of death, myocardial infarction, revascularization, and resource use. The study was approved by the institutional review boards of all sites. Data Collection and Processing Structured data collection was performed prospectively at the index visit in accordance with standardized reporting guidelines9 and key definitions10 and included demographic and clinical characteristics, the ECG, treatment, diagnostic testing, and disposition. Subjects were contacted by telephone at 30 days and 1 year after ED presentation and queried about myocardial infarction, ED visits or hospitalizations, revascularization, cardiac testing (coronary CT angiography, stress testing, echocardiography, and catheterization), cardiologist visits, and cardiac medication use. Medical record review was performed for all potential cardiac hospitalizations, cardiac diagnostic tests, myocardial infarction, and all deaths. If the subjects or secondary contacts were unavailable, records at the presenting and neighboring hospitals were reviewed for repeated visits. When these methods failed to provide survival information, we searched the Social Security Death Master File (http://www.ssdmf.com) for vital status (January 14, 2013). An independent committee adjudicated all potential myocardial infarctions and cardiac-related deaths, using standard definitions.10,11 Primary Data Analysis The trial was powered to test the principal hypothesis that the 30-day major adverse cardiac event rate (including myocardial infarction and cardiac death) among patients found not to have significant coronary artery disease on coronary CT angiography exceeds 1%. Given expectations Volume

-,

no.

-

:

-

2015

Hollander et al

One-Year Outcomes of ACRIN Trial

about the rate of coronary artery disease, cardiac events, and attrition according to previous studies, a sample size of 910 participants in the coronary CT angiography arm provided power of at least 90% to reject the null hypothesis, using an exact 1-sided test of level a¼0.05. Given the low event rate anticipated in the population being studied, practical considerations meant that we could not power the study to compare 1-year safety between the coronary CT angiography and traditional care arms. We compared rates of major adverse cardiac event, revascularization, and resource use (including repeated ED visits, hospitalizations, cardiac testing, and cardiac medication use) at 1 year between arms. Exact confidence intervals (CIs) for the difference in proportions between arms were computed. In addition, the 1-year safety rate in terms of major adverse cardiac event for the coronary CT angiography–negative subset was calculated, along with the corresponding exact binomial CI. Statistical computations were conducted with SAS (version 9.3; SAS Institute, Inc., Cary, NC) and R (version 3.1.0; R project, http://www.r-project.org). RESULTS We enrolled 1,392 subjects; 23 were removed according to predetermined criteria, including renal insufficiency not known at randomization, subject withdrawal, or having received a CT to exclude pulmonary embolism because of an elevated D-dimer level. In addition, although index data were collected, 1 subject withdrew from further follow-up after the 30-day contact, leaving a cohort of 1,368 for analysis (Figure). One-year vital status information was available from direct participant or proxy contact for 1,285 patients (94%), medical record review for 46 subjects (3.4%), and the Social Security Death Index for 37 subjects (2.7%). Follow-up rates in each arm were not different. Demographic and clinical characteristics of enrolled patients and those lost to follow-up are show in Table E1, available online at http://www.annemergmed.com. There were no differences between groups with respect to overall or cardiac-related death, myocardial infarction, or revascularization rates from index presentation through 1 year (Table 1) or from ED or hospital discharge through 1 year (Table 2). Details concerning the 5 observed deaths through 1 year and the adjudicated cause of death are shown in Table 3. There was only 1 cardiac death in the entire cohort. No difference between arms was observed in regard to mortality. There were 2 deaths overall in the coronary CT angiography arm and 3 in the traditional care arm (0.2% versus 0.7%; difference –0.4%; 95% CI –6.0% to 5.2%). One death in the coronary CT angiography arm and none in the traditional care arm were adjudicated as cardiac Volume

-,

no.

-

:

-

2015

related (0.1% versus 0%; difference 0.1%; 95% CI –5.5% to 5.7%). All deaths occurred between 30 days and 1 year after enrollment. There was no difference between arms in the rate of myocardial infarction from index presentation through 1 year (1.3% versus 1.1%; difference 0.1%; 95% CI –5.6% to 5.9%) (Table 1), or from ED or hospital discharge through 1 year (0.2% versus 0.5%; difference –0.2%; 95% CI –5.9% to 5.5%) (Table 2). One myocardial infarction occurred in the coronary CT angiography arm between 30 days and 1 year and none in the traditional care arm (0.1% versus 0%; difference 0.1%; 95% CI –5.6% to 5.8%); however, this patient did not undergo coronary CT angiography during the index visit because of large body habitus. In combining cardiac death and myocardial infarction, there was likewise no difference between arms in the rate of major adverse cardiac events from index presentation through 1 year (1.4% versus 1.1%; difference 0.3%; 95% CI –5.5% to 6.0%) (Table 1), from ED or hospital discharge through 1 year (0.3% versus 0.5%; difference –0.1%; 95% CI –5.8% to 5.6%) (Table 2), or from 30 days to 1 year (0.2% versus 0%; difference 0.2%; 95% CI –5.5% to 6.0%). One of 640 subjects with a negative coronary CT angiography result during the index visit had a cardiac death and none had a myocardial infarction within 1 year of presentation (major adverse cardiac event rate 0.16%; 95% CI 0.004% to 0.87%) (see Table 3 for details). Finally, there was no statistically significant difference in the rate of revascularization between arms from index presentation through 1 year (2.9% versus 1.6%; difference 1.3%; 95% CI –4.4% to 7.0%) (Table 1) or from ED or hospital discharge through 1 year (0.5% versus 0.9%; difference –0.4%; 95% CI –6.2% to 5.3%) (Table 2). There were 3 revascularizations in the coronary CT angiography arm from 30 days to 1 year compared with 1 in the traditional care arm (0.3% versus 0.2%; difference 0.1%; 95% CI –5.6% to 5.8%). There was no statistically significant difference in the use of any additional cardiac testing between the coronary CT angiography and traditional care arms from ED or hospital discharge through 1 year (13% versus 13%; difference –0.4%; 95% CI –6.2% to 5.5%). There were no differences in any individual measure of resource use between groups, including rate of invasive angiography or noninvasive cardiac testing (echocardiography, stress testing with or without imaging, and coronary CT angiography) (Table 4). There was also no difference in the rate of repeated cardiac ED visits, cardiac hospitalization, or cardiologist office visits. Finally, there were no statistically significant differences between the 2 arms in the usage rates of aspirin, thienopyridines, or statins at 1 year. Annals of Emergency Medicine 3

Hollander et al

One-Year Outcomes of ACRIN Trial

Figure. Study flow chart including 1-year follow-up rates. CCTA, Coronary CT angiography; AMI, acute myocardial infarction.

LIMITATIONS The main limitation of this study is that, given the low event rates in this population, we were not able to power it to detect a difference in safety between arms. However, we were able to demonstrate an acceptable 1-year safety threshold for patients with a negative coronary CT angiography result. Additionally, the trial was not powered specifically to detect differences in the resource use rates between arms; however, although we may have missed small differences, coronary CT angiography does not appear to significantly increase resource use. We did not prespecify an analysis according to risk (ie, Thrombolysis in Myocardial Infarction score), but because the whole group 4 Annals of Emergency Medicine

was effectively free from adverse events, the same would be true within any subgroup included in this study. There is a possibility of underreporting of events at 1 year, but there is no particular reason to believe it would be more or less likely in one group of patients. Finally, we cannot be certain that patients lost to follow-up were not more likely to experience adverse events than those we followed; however, because there was not a difference between groups in the rate of lost to follow-up and the 2 groups were well matched at entry, it is unlikely that there would be a much larger number of events in one of these groups than the other. We believe the bias would be in the direction of the traditional care group having worse outcomes because Volume

-,

no.

-

:

-

2015

Hollander et al

One-Year Outcomes of ACRIN Trial

Table 1. Event rates in coronary CT angiography and traditional care arms from randomization through 1 year.

Event

CCTA Traditional Care (n[907) (%) (n[461) (%)

Difference (95% CI), CCTA-Traditional Care, %

All-cause mortality Cardiac death AMI MACE Revascularization

2/907 1/906 11/870 12/870 25/872

–0.4 0.1 0.1 0.3 1.3

(0.2) (0.1) (1.3) (1.4) (2.9)

3/461 (0.7) 0/460 5/444 (1.1) 5/443 (1.1) 7/444 (1.6)

(–6.0 (–5.5 (–5.6 (–5.5 (–4.4

to to to to to

5.2) 5.7) 5.9) 6.0) 7.0)

MACE, major adverse cardiac event (composite of cardiac death and AMI).

more patients in the coronary CT angiography group were known to be free of disease and fewer patients in the traditional care group received testing during the index visit. DISCUSSION In the initial report of the ACRIN PA 4005 study,2 we showed that event rates during the index visit and at 30 days were the same for patients with low- to intermediaterisk chest pain in the ED who were randomized to a pathway including coronary CT angiography versus a traditional care pathway. This report extends these findings, demonstrating no difference in event rates from index presentation, ED or hospital discharge, or 30 days through 1 year. We expected to find a decrease in resource use in the coronary CT angiography group. Nerenberg et al12 and Shaver et al13 demonstrated that relative to patients without any previous testing, those known to have negative stress test results were just as likely to return to the ED with chest pain, to be admitted to the hospital, and to receive cardiac catheterization. We expected that the proven absence of coronary artery disease on coronary CT angiography would make these events less likely to happen. However, we found that resource use was the same for the 2 groups (coronary CT angiography and traditional care) through 1 year, including similar rates of revascularization, cardiac catheterization, and noninvasive cardiac testing, as well as ED and cardiologist visits and medication use. This Table 2. Event rates in coronary CT angiography and traditional care arms from ED or hospital discharge through 1 year.

CCTA Traditional Care (n[907) (%) (n[461) (%)

Event All-cause mortality Cardiac death AMI MACE Revascularization

Volume

-,

no.

-

2/907 1/906 2/870 3/870 4/870

:

-

(0.2) (0.1) (0.2) (0.3) (0.5)

2015

3/461 (0.7) 0/460 2/443 (0.5) 2/442 (0.5) 4/443 (0.9)

Difference (95% CI), CCTA-Traditional Care, % –0.4 0.1 –0.2 –0.1 –0.4

(–6.0 (–5.5 (–5.9 (–5.8 (–6.2

to to to to to

5.2) 5.7) 5.5) 5.6) 5.3)

Table 3. Deaths in the coronary CT angiography and traditional care arms from index visit through 1 year, with results of adjudication committee determination and specific cause of death. Study Arm

CCTA Result

Death Cardiac Related

Specific Cause of Death (If Known)

CCTA CCTA

Negative Negative

Unknown Yes

Traditional care Traditional care Traditional care

N/A

No

N/A

Unknown

Unknown Ventricular fibrillation arrest/ sudden cardiac death. Patient reported history of cocaine use during index visit. Gunshot wound to head, suicide Metastatic breast cancer

N/A

No

Non-Hodgkin’s lymphoma

N/A, Not applicable (CCTA was not administered).

finding is in contrast with a reported increase in subsequent cardiac testing, catheterization, revascularization, and total cost of care among Medicare beneficiaries undergoing nonemergency coronary CT angiography versus stress testing.14 This difference is likely related to the higher pretest probability of coronary artery disease in the ambulatory Medicare beneficiary population studied than the typically younger ED acute chest pain population (74 versus 49 years in our trial). At a high enough pretest probability of disease, it is possible that coronary CT angiography will result in further testing or increased rates of revascularization. Although meta-analyses4,5 revealed an increase in revascularizations and possibly catheterization rates in the initial reports of multicenter trials of coronary CT angiography versus traditional care, our findings suggest that this difference does not persist beyond the index visit. In our trial, resource use was not increased at any point (index visit, 30 days, or 1 year). There was, however, a lower rate of negative catheterization results (defined as no stenosis
50%) for coronary CT angiography versus traditional care (24% versus 56% during the index visit and 29% versus 53% within 30 days), suggesting more appropriate use of catheterization in the coronary CT angiography arm.2 This finding is similar to that of the Prospective Multicenter Imaging Study for Evaluation of Chest Pain, in which symptomatic outpatients who received coronary CT angiography were less likely to have catheterizations without obstructive disease than was seen with functional testing.15 The difficulty in determining the functional significance of an anatomic stenosis at coronary CT angiography is being addressed by 2 new technologies that could find application in the evaluation of ED chest pain patients. Perfusion CT adds an evaluation of myocardial perfusion at rest or under pharmacologic stress to the anatomic Annals of Emergency Medicine 5

Hollander et al

One-Year Outcomes of ACRIN Trial Table 4. Resource use and reported medication use from ED or hospital discharge through 1 year. Outcome ED visit Hospital admission Cardiologist office visit Testing CCTA Stress testing without imaging Stress testing with imaging Catheterization Echocardiogram Medication use Aspirin Thienopyridines Statins

CCTA (%)

Traditional Care (%)

Difference (95% Exact CI), CCTA–Traditional Care, %

305/852 (36) 137/843 (16) 148/821 (18)

166/438 (38) 74/432 (17) 53/423 (13)

–2.1 (–7.9 to 3.7) –0.9 (–6.7 to 4.9) 5.5 (–0.4 to 11.3)

2/822 19/827 60/824 24/829 49/824

(0.2) (2) (7) (3) (6)

283/843 (34) 37/830 (4) 198/835 (24)

depiction of coronary anatomy and disease provided by coronary CT angiography.16 Results of the first multicenter trial of stress perfusion CT demonstrated a significant improvement in accuracy over coronary CT angiography alone compared with a criterion standard combining greater than or equal to 50% stenosis at catheter angiography and presence of a perfusion defect at singlephoton emission computerized tomography (SPECT)– myocardial perfusion imaging.17 One study has evaluated the performance of rest perfusion CT in ED chest pain patients,18 demonstrating improved positive predictive value for prediction of SPECT–myocardial perfusion imaging defects compared with coronary CT angiography alone. A measure analogous to catheter-based fractional flow reserve can be derived from coronary CT angiography data with computational fluid dynamics under simulated adenosine hyperemia conditions.19 Several multicenter trials of this technology have been published, including a recent study that demonstrated improved accuracy (particularly specificity) compared with coronary CT angiography alone when compared with catheter-based fractional flow reserve.20 Application of these techniques to evaluation of ED chest pain may further decrease downstream resource use, particularly rates of catheterization and revascularization. In contrast to the initial report of the ACRIN PA trial,2 which had a 4% rate of acute coronary syndrome and 7% rate of coronary disease, ROMICAT II3 had an 8% rate of acute coronary syndrome; however, the majority (nearly 70%) of these patients received a diagnosis of non acute myocardial infarction acute coronary syndrome, which is a clinical diagnosis without necessarily having objective findings. It is possible the slightly older age (54 versus 50 years) and increased enrollment of men (53% versus 47%) can also explain this difference. Even with a possibly higher-risk group of patients in ROMICAT II, there were 6 Annals of Emergency Medicine

5/422 2/424 33/422 13/423 26/425

(1) (0.5) (8) (3) (6)

159/433 (37) 12/430 (3) 78/431 (18)

–0.9 1.8 –0.5 –0.2 –0.2

(–6.8 (–4.0 (–6.4 (–6.0 (–6.0

to to to to to

4.9) 7.7) 5.3) 5.7) 5.7)

–3.2 (–8.9 to 2.7) 1.7 (–4.2 to 7.5) 5.6 (–0.2 to 11.4)

no undetected cases of acute coronary syndrome within 28 days. We demonstrated a 1-year event rate of less than 1% in patients with a negative coronary CT angiography result. This study provides multicenter support for findings of smaller studies. In a single-center study of 481 patients with low- to intermediate-risk chest pain in the ED, without significant coronary artery disease by coronary CT angiography, Hollander et al6 reported no acute myocardial infarction or revascularizations within 1 year. One subject with a normal coronary CT angiography result died of an unknown cause; assuming that this death was cardiac, the observed mortality rate was 0.2%. The ROMICAT I trial,21 a single-center observational study in which 368 patients received coronary CT angiography in addition to traditional care, but treating physicians were blinded to the coronary CT angiography results, reported no events at 2 years in patients without any plaque or stenosis on coronary CT angiography. There was a 4.6% event rate (including revascularization, as well as acute myocardial infarction and cardiac death) in patients with nonobstructive coronary artery disease (plaque or stenosis <50%); however, only 1 patient had an event between 30 days and 2 years (1.2%). In the Coronary Computed Tomography for Systematic Triage of Acute Chest Pain Patients to Treatment trial,1 in which ED chest pain patients were randomized to coronary CT angiography or SPECT–myocardial perfusion imaging, the 6-month event rate for patients who had normal or near-normal coronary CT angiography results (<25% stenosis) was 0.8% (2/268 patients, 2 revascularizations, zero acute myocardial infarction/deaths). These studies provide evidence of the durability of a negative ED coronary CT angiography result for predicting low risk of future events and mirror results obtained in the general population of patients undergoing coronary CT angiography. For example, in the Coronary CT Angiography Evaluation for Clinical Outcomes registry Volume

-,

no.

-

:

-

2015

Hollander et al

One-Year Outcomes of ACRIN Trial

of more than 14,000 patients, the annualized all-cause mortality rates for patients without coronary artery disease and those with nonobstructive disease were 0.36% and 1.14%, respectively.22 More directly, Chang et al23 reported on findings of repeated imaging performed after a previous ED coronary CT angiography. At a median of 27 months between examinations, none of the 29 patients with a maximal stenosis less than 50% on the initial study progressed to a greater than 50% stenosis at follow-up. Some have editorialized that testing may not be necessary at all in low-risk patients,24-26 noting that testing has not been shown to decrease event rates or improve outcomes in these patients compared with accelerated biomarker-only protocols. To our knowledge, there have been no randomized trials comparing testing versus no-testing strategies; therefore, we cannot know whether the assertion is true, and such a trial would require a prohibitive number of subjects, given the low event rates in this population. One recent report examined private insurance claims data for a large cohort of patients who had an ED chest pain evaluation.27 This study demonstrated no difference in the rate of hospitalization for myocardial infarction at 7 and 190 days between patients who had an imaging test during the ED visit and those who had no testing, but those who had any imaging test had higher rates of catheterization, revascularization, and further noninvasive testing. The authors concluded that deferral of noninvasive testing for these patients may be reasonable. However, patients in the testing group were older and had more comorbidities than those in the no-testing group, so the significance of these findings is unknown. Also, this approach does not deal effectively with the problem of recidivism and does not take advantage of the opportunity for long-term coronary artery disease risk assessment and intervention (using statins) in patients not having an acute coronary syndrome. One concern is whether performing coronary CT angiography on low- to intermediate-risk patient populations leads to overtesting, but we did not find that to be the case in this study. Finally, current American College of Cardiology/ American Heart Association practice guidelines recommend testing for coronary artery disease after exclusion of myocardial infarction.28 It is critical for providers to integrate test results with clinical judgment. The mere finding of coronary disease on coronary CT angiography does not mean that the coronary disease is the cause of the symptoms that led to the test. Similarly, appearing to have reversible ischemia on a stress test does mean the patient actually has ischemic heart disease. Many of these tests can be falsely positive. In conclusion, the use of a coronary CT angiography based strategy for evaluation of ED patients with lowto intermediate-risk chest pain does not lead to increased Volume

-,

no.

-

:

-

2015

event rates or resource use through 1 year, and a negative coronary CT angiography result confers a less than 1% event rate at 1 year. Along with the improvements in efficiency demonstrated by multiple randomized controlled trials, our results provide further motivation for implementation of coronary CT angiography–based strategies for evaluation of these patients. The authors acknowledge the physicians, nurses, technicians, and research assistants at the participating institutions, as well as the ACRIN staff who supported the trial at both ACRIN headquarters in Philadelphia, PA, and the Biostatistics and Data Management Center at Brown University, Providence, RI. Without the diligent efforts of these many individuals, this study would not have been possible. Supervising editor: Deborah B. Diercks, MD Author affiliations: From the Department of Emergency Medicine, Sidney Kimmel Medical College of Thomas Jefferson University, Philadelphia, PA (Hollander, Chang); the Department of Radiology, Perelman School of Medicine of the University of Pennsylvania, Philadelphia, PA (Litt); the Center for Statistical Sciences (Gatsonis, Greco, Snyder) and Department of Biostatistics (Gatsonis), Brown University School of Public Health, Providence, RI; the Department of Emergency Medicine, Wake Forest School of Medicine, WinstonSalem, NC (Miller); and the Department of Radiology, Penn State Hershey Medical Center, Hershey, PA (Singh). Author contributions: JEH, CG, BSS, and HIL designed the study. All authors performed data collection, interpretation of results, and study oversight at their respective sites. CG, EMG, and BSS were responsible for data analysis. JEH and HIL were responsible for preparing the article, and all authors critically reviewed it. JEH takes responsibility for the paper as a whole. Funding and support: By Annals policy, all authors are required to disclose any and all commercial, financial, and other relationships in any way related to the subject of this article as per ICMJE conflict of interest guidelines (see www.icmje.org). The authors have stated that no such relationships exist and provided the following details: This project is funded, in part, under a grant with the Pennsylvania Department of Health. The Department specifically declaims responsibility for any analyses, interpretations or conclusions. Additional funding was provided by the ACR Fund for Imaging Innovation. Publication dates: Received for publication July 21, 2015. Revision received August 24, 2015. Accepted for publication September 14, 2015. Trial registration number: NCT00933400

REFERENCES 1. Goldstein JA, Chinnaiyan KM, Abidov A, et al; for the CT-STAT Investigators. The CT-STAT (Coronary Computed Tomographic Angiography for Systematic Triage of Acute Chest Pain Patients to Treatment) trial. J Am Coll Cardiol. 2011;58:1414-1422.

Annals of Emergency Medicine 7

Hollander et al

One-Year Outcomes of ACRIN Trial 2. Litt HI, Gatsonis C, Snyder B, et al. CT angiography for safe discharge of patients with possible acute coronary syndromes. N Engl J Med. 2012;366:1393-1403. 3. Hoffmann U, Truong QA, Schoenfeld DA, et al; ROMICAT-II Investigators. Coronary CT angiography versus standard evaluation in acute chest pain. N Engl J Med. 2012;367:299-308. 4. D’Ascenzo F, Cerrato E, Biondi-Zoccai G, et al. Coronary computed tomographic angiography for detection of coronary artery disease in patients presenting to the emergency department with chest pain: a meta-analysis of randomized clinical trials. Eur Heart J Cardiovasc Imaging. 2013;14:782-789. 5. Hulten E, Pickett C, Bittencourt MS, et al. Outcomes after coronary computed tomography angiography in the emergency department: a systematic review and meta-analysis of randomized, controlled trials. J Am Coll Cardiol. 2013;61:880-892. 6. Hollander JE, Chang AM, Shofer FS, et al. One-year outcomes following coronary computerized tomographic angiography for evaluation of emergency department patients with potential acute coronary syndrome. Acad Emerg Med. 2009;16:693-698. 7. Schlett CL, Banerji D, Siegel E, et al. Prognostic value of CT angiography for major adverse cardiac events in patients with acute chest pain from the emergency department: 2-year outcomes of the ROMICAT trial. JACC Cardiovasc Imaging. 2011;4: 481-491. 8. Antman EM, Cohen M, Bernink PM, et al. The TIMI risk score for unstable angina/non–ST elevation MI: a method for prognostication and therapeutic decision making. JAMA. 2000;284:835-842. 9. Hollander JE, Blomkalns AL, Brogan GX, et al. Standardized reporting guidelines for studies evaluating risk stratification of emergency department patients with potential acute coronary syndromes. Ann Emerg Med. 2004;44:589-598. 10. Cannon CP, Battler A, Brindis RG, et al. American College of Cardiology key data elements and definitions for measuring the clinical management and outcomes of patients with acute coronary syndromes. A report of the American College of Cardiology Task Force on Clinical Data Standards (Acute Coronary Syndromes Writing Committee). J Am Coll Cardiol. 2001;38:2114-2130. 11. Thygesen K, Alpert JS, White HD, et al. Universal definition of myocardial infarction. Circulation. 2007;116:2634-2653. 12. Nerenberg RH, Shofer FS, Robey JL, et al. Impact of a negative prior stress test on emergency physician disposition decision in ED patients with chest pain syndromes. Am J Emerg Med. 2007;25:39-44. 13. Shaver KJ, Marsan RJ, Sease KL, et al. Impact of a negative evaluation for underlying coronary artery disease on 1 year resource utilization for patients admitted with potential acute coronary syndromes. Acad Emerg Med. 2005;11:1272-1277. 14. Shreibati J, Baker LC, Hlatky MA. Association of coronary CT angiography or stress testing with subsequent utilization and spending among Medicare beneficiaries. JAMA. 2011;306:2128-2136. 15. Douglas PS, Hoffman U, Patel MR, et al; for the PROMISE Investigators. Outcomes of anatomical versus functional testing for coronary artery disease. N Engl J Med. 2015;372:1291-1300. 16. Rossi A, Merkus D, Klotz E, et al. Stress myocardial perfusion: imaging with multidetector CT. Radiology. 2014;270:25-46. 17. Rochitte CE, George RT, Chen MY, et al. Computed tomography angiography and perfusion to assess coronary artery stenosis causing perfusion defects by single photon emission computed tomography: the CORE320 Study. Eur Heart J. 2014;35: 1120-1130. 18. Feuchtner GM, Plank F, Pena C, et al. Evaluation of myocardial CT perfusion in patients presenting with acute chest pain to the emergency department: comparison with SPECT–myocardial perfusion imaging. Heart. 2012;98:1510-1517. 19. Grunau GL, Min JK, Leipsic J. Modeling of fractional flow reserve based on coronary CT angiography. Curr Cardiol Rep. 2013;15: 336-342.

8 Annals of Emergency Medicine

20. Nørgaard BL, Leipsic J, Gaur S, et al. Diagnostic performance of noninvasive fractional flow reserve derived from coronary computed tomography angiography in suspected coronary artery disease. J Am Coll Cardiol. 2014;63:1145-1155. 21. Hoffmann U, Bamberg F, Chae CU, 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;53:1642-1650. 22. Chow BJ, Small G, Yam Y, et al; CONFIRM Investigators. Incremental prognostic value of cardiac computed tomography in coronary artery disease using CONFIRM: Coronary Computed Tomography Angiography Evaluation for Clinical Outcomes: An International Multicenter Registry. Circ Cardiovasc Imaging. 2011;4:463-472. 23. Chang AM, Ginty CT, Litt HI, et al. Coronary artery disease progression in patients without significant stenosis on coronary computed tomographic angiography. Am J Emerg Med. 2012;30:2015-2020. 24. Kosowsky JM. Approach to the ED patient with “low-risk” chest pain. Emerg Med Clin North Am. 2011;29:721-727, vi. 25. Redberg RF. Coronary CT angiography for acute chest pain. N Engl J Med. 2012;367:375-376. 26. Redberg RF. Stress testing in the emergency department: not which test but whether any test should be done. JAMA Intern Med. 2015;175:436. 27. Foy AJ, Liu G, Davidson WR Jr, et al. Comparative effectiveness of diagnostic testing strategies in emergency department patients with chest pain: an analysis of downstream testing, interventions, and outcomes. JAMA Intern Med. 2015;175:428-436. 28. Amsterdam EA, Kirk JD, Bluemke DA, et al; American Heart Association Exercise, Cardiac Rehabilitation, and Prevention Committee of the Council on Clinical Cardiology, Council on Cardiovascular Nursing, and Interdisciplinary Council on Quality of Care and Outcomes Research. Testing of low-risk patients presenting to the emergency department with chest pain: a scientific statement from the American Heart Association. Circulation. 2010;122:1756-1776.

APPENDIX Hospital of the University of Pennsylvania: Site Investigators: Judd E. Hollander, MD, Harold I. Litt, MD, PhD; Research Coordinators: Emily Barrows, Jeffrey Le, Shannon Marcoon, Julie Pitts, RN, Scott Steingall, RT Pennsylvania State University Medical Center at Hershey: Site Investigators: James M. Leaming, MD, Harjit Singh, MD, Michelle A. Fischer, MD, Steven Ettinger, MD, Carlos Jamis-Dow, MD, Kevin Moser, MD; Research Coordinators: Swati Shah, Kevin Gardner, RN, Russell Dicristina, Susan Oskorus Penn Presbyterian Medical Center: Site Investigators: Laurence Gavin, MD, Anna Marie Chang, MD, Judd E. Hollander, MD, Harold I. Litt, MD, PhD; Research Coordinators: Christopher Decker, Michael Green, Katie O’Conor, Angela Roach, Scott Steingall, RT, Kristy Walsh, Max Wayne Wake Forest University: Site Investigators: J. Jeffrey Carr, MD, MSc, Daniel W. Entrikin, MD, Kim Askew, MD, James W. Hoekstra, MD, Simon Mahler, MD, Chadwick D. Miller, MD, MS; Research Coordinators: Volume

-,

no.

-

:

-

2015

Hollander et al

One-Year Outcomes of ACRIN Trial

Denise Boyles, Stephanie Bradshaw, Mark Collin, Erin Harper, Lisa Hinshaw, MS, Jane Kilkenny, Megan Koonts, Lori Triplett, RN University of Pittsburgh Medical Center: Site Investigators: Charissa B. Pacella, MD, Joan M. Lacomis, MD, Christopher R. Deible, MD, PhD; Research Coordinators: Sara Vandruff, Barbara Early, Tina Vita, Dawn McBride Brown University: Biostatistical and research design support from Constantine Gatsonis, PhD, Brad Snyder, MS, Sanaa Boudhar, MS, Patricia Fox, MS, Erin Greco, MS

Volume

-,

no.

-

:

-

2015

Data and Safety Monitoring Board: David Bluemke, MD, PhD (Chair), National Institutes of Health, Bethesda, MD; Todd A. Alonzo, PhD, University of Southern California, Arcadia, CA; Jon F. Merz, MBA, JD, PhD, University of Pennsylvania, Philadelphia, PA; Herbert Y. Kressel, MD, Beth Israel Deaconess Medical Center, Boston, MA Adjudication Committee: W. Frank Peacock, MD, Cleveland Clinic, OH; Robert Hendel, MD, University of Miami, FL

Annals of Emergency Medicine 9

Hollander et al

One-Year Outcomes of ACRIN Trial Table E1. Demographic, historical, and presenting characteristics by contact status. Characteristic Age, mean (SD; range), y Sex Male Female Ethnicity Hispanic or Latino Not Hispanic or Latino Unknown Race† American Indian or Alaskan Native Asian Black or African American Native Hawaiian or other Pacific Islander White Unknown Cardiac history and risk factors† Hypertension Hypercholesterolemia Family history of CAD Diabetes mellitus Current tobacco use Cocaine use in last week Myocardial infarction Congestive failure Arrhythmias Pulse at presentation, beats/min
80 60–79 <60 ECG findings Normal Nonspecific Early repolarization Abnormal, not diagnostic Ischemia, old/unchanged ECG Ischemia, not known to be old Suggestive of MI TIMI risk score 0 1 2 3

Participant/Proxy Contact (N[1,285), No. (%)

No Participant Contact (N[83), No. (%)*

49 (9.10; 30–83)

45 (8.35; 31–65)

588 (45.76) 697 (54.24)

57 (68.67) 26 (31.33)

30 (2.33) 1,225 (95.33) 30 (2.33)

2 (2.41) 79 (95.18) 2 (2.41)

10 18 772 2 480 12

(0.78) (1.40) (60.08) (0.16) (37.35) (0.93)

1 (1.20) 0 40 (48.19) 0 42 (50.60) 1 (1.20)

659 353 367 187 406 62 16 19 25

(51.28) (27.47) (28.56) (14.55) (31.60) (4.82) (1.25) (1.48) (1.95)

35 14 27 7 40 7

(42.17) (16.87) (32.53) (8.43) (48.19) (8.43) 0 0 2 (2.41)

718 (55.88) 520 (40.47) 47 (3.66)

50 (60.24) 32 (38.55) 1 (1.20)

827 301 34 88 14 17 2

(64.46) (23.46) (2.65) (6.86) (1.09) (1.33) (0.16)

55 17 4 4 3

652 459 173 1

(50.74) (35.72) (13.46) (0.08)

41 (49.40) 32 (38.55) 10 (12.05) 0

(66.27) (20.48) (4.82) (4.82) (3.61) 0 0

CAD, Coronary artery disease; MI, myocardial infarction; TIMI, Thrombolysis in Myocardial Infarction. *Includes patients who had medical record review but were not directly contacted, as well as those who had only a death index search (n¼37). † Participants may have reported more than one category/characteristic; percentages may not sum to 100.

9.e1 Annals of Emergency Medicine

Volume

-,

no.

-

:

-

2015