Coronary CT Angiography to Rule Out Acute Coronary Syndromes in the Emergency Department

Abstract & commentary

By Andrew J. Boyle, MBBS, PhD, Assistant Professor of Medicine, Interventional Cardiology, University of California, San Francisco. Dr. Boyle reports no financial relationships relevant to this field of study.

Source: 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;53:1642-1650.

Patients presenting to the emergency department (ED) with acute chest pain are a significant portion of our health care budget. Current recommendations for the assessment and management of these patients involve extended periods of observation for repeated biomarkers and electrocardiograms (ECG). This often results in hospital admission to "rule out" myocardial infarction. Any advance in the speed or accuracy of diagnosis of the cause of chest pain, or to rule out myocardial ischemia as the cause, would be a significant clinical advance. Coronary computed tomography (CT) angiography (CCTA) has emerged as an accurate diagnostic imaging modality in suitable patients with stable coronary artery disease (CAD). However, its role in the triage of patients in the ED with acute chest pain has not been fully defined. Accordingly, Hoffman et al conducted a prospective, observational cohort study to assess the utility of CCTA in patients presenting with acute chest pain, who were being admitted and observed with low- to intermediate-risk for acute coronary syndromes (ACS).

Hoffman et al enrolled adult patients presenting to the ED with chest pain of at least five minutes duration, who were being admitted for observation for possible ACS, who were in sinus rhythm and were able to perform an adequate breath hold for the CCTA. They excluded patients with known coronary artery disease, elevated biomarkers, or ECG changes. Patients who were not suitable for CCTA also were excluded — those with renal impairment, contrast allergy, thyroid disorders, hemodynamic instability, or taking metformin. Patients underwent CCTA prior to admission, and the treating physicians were not informed of the results. The diagnosis of ACS was made by standard practice, without knowledge of the CCTA findings. Hoffman et al screened 1,869 patients and enrolled 368 in the study, mean age 53 ± 12 years; 61% were male. Mean BMI was 29 6 kg/m2, 11% were diabetic, 39% had hypertension, 49% current or former smokers, and 37% had measured dyslipidemia. CCTA was performed in standard fashion, with two blinded reviewers reporting the results as normal, presence of plaque, or presence of > 50% stenosis. When there was discrepancy, a third reviewer was consulted.

ACS was diagnosed in 31 of 368 patients (8.4% of the cohort; unstable angina in 23, myocardial infarction [MI] in 8). All patients with ACS had coronary plaque (100% sensitivity and 100% negative predictive value), whereas 24 of 31 also had coronary stenosis > 50% (77% sensitivity and 98% negative predictive value). In the subgroup of patients with myocardial infarction, plaque was present in all (100% sensitivity and 100% negative predictive value) and stenosis > 50% was present in five of eight patients (63% sensitivity and 99% negative predictive value). No patient without coronary plaque developed ACS during that hospitalization or over the ensuing six months. Despite the excellent sensitivity and negative predictive value of coronary plaque for ACS, the specificity and positive predictive value were low to moderate because many patients had plaque but did not have ACS (specificity 54% and positive predictive value 17%). Importantly, in patients who did not have ACS, there were no major adverse cardiac events (MACE) at six months.

Among the 185 subjects with any coronary plaque detected, those who were diagnosed with ACS had a greater burden of coronary plaque, with more coronary artery segments involved (7.2 vs. 4.2 segments; p < 0.0001), they had more calcified plaque (6.5 vs. 3.6 segments; p < 0.0001) and more non-calcified plaque (3.6 vs. 1.8 segments; p < 0.0001) than patients who were not diagnosed with ACS. The type of plaque detected by CCTA did not influence the likelihood of ACS.

The presence of coronary stenosis > 50% was not as sensitive for detecting ACS as the presence of any coronary plaque. Sensitivity was 77% and negative predictive value was 98%, because there were seven patients without stenosis on CCTA who were actually found to have ACS. The specificity of coronary stenosis >50% for detecting ACS was lower in those over 65 (58% vs. 91%; p < 0.0001), and these patients had a higher prevalence of coronary calcium (84% vs. 39%; p < 0.0001). In logistic regression analysis, after adjustment for age sex, and TIMI risk score, each additional segment of plaque was associated with a 28% increase in having ACS (p < 0.0001), and the presence of stenosis >50% increased the odds ratio of having ACS by over eleven-fold (OR 11.69; p < 0.0001). The authors conclude that both plaque and stenosis detected by CCTA predict ACS independent of risk factors or TIMI risk score. Fifty percent of patients with chest pain and low to intermediate risk are free of CAD and have no ACS. Given the large number of such patients, early CCTA may significantly improve patient management in the ED.


Patients presenting with ACS and high-risk features are treated with aggressive medical therapy and often with an early invasive strategy. But patients presenting with low- or intermediate-risk ACS represent a more heterogeneous group and the optimal management strategy is less clear. Inappropriate discharges after missed ACS can result in MI or death. Thus, clinicians err on the side of caution and often admit low- to intermediate-risk patients for observation and serial ECG and biomarkers. CCTA in the ED holds promise to add incremental value to our current risk stratification protocols. This study by Hoffmann et al confirms the potential of CCTA to aid in decision-making in patients with low to intermediate risk of ACS. In their study, 50% of patients had no coronary disease by CCTA: none of these patients were found to have ACS on admission and none had MACE in the 6-month follow-up period. This may have important implications for improving work-flow in the ED, reducing inappropriate admissions and reducing health care expenditure. However, the results presented herein, although exciting and provocative, require confirmation in other centers before becoming part of the standard of care.

This study design allows for unbiased assessment of the performance of the CCTA, because the results were not communicated to the clinicians and, therefore, were not used as part of the clinical decision-making process. Thus standard diagnostic measures could be compared to the CCTA results, rather than using the CCTA results as part of the diagnosis. The absence of MACE during the six-month follow-up for patients without coronary artery disease is reassuring and lends weight to the use of CCTA as a decision-making tool in the ED. However, there are some limitations to the study. Firstly, many patients were excluded, including those who presented outside business hours, those with known coronary artery disease, and those with renal impairment. Secondly, elderly patients were under-represented in this cohort. These patient groups make up a large percentage of those presenting with chest pain. Thirdly, this was a single-center study with physicians highly experienced in CCTA. Many centers are not so experienced in interpreting this evolving technology. These factors may mean the results of this study are not generalizable to all centers. Finally, only 92% of patients were evaluable at 6-month follow-up, and therefore it is possible that some clinical events were missed. Prior to changing standard practice, these results should be confirmed in multi-center randomized studies, with rigorous follow-up, that are powered to detect clinical end-points.