Diagnostic Performance of Coronary CTA in Intermediate-to-High-Risk Patients for Suspected Acute Coronary Syndrome

Results From an Emergency Department Registry

Coronary computed tomography angiography (CTA) is an excellent tool for ruling out coronary artery disease (CAD) in low-risk patients (1). High-sensitivity troponin assays have proven safety and efficacy in ruling out large numbers of patients with a low risk for acute coronary syndrome (ACS), leading to no additional need for imaging, but in parallel resulting in more patients with elevated biomarkers. In an aging population with medical therapies improving longevity, more patients with higher risk for ACS will present to the emergency department (ED).

We hypothesized that CTA may offer a safe and effective tool to detect or exclude obstructive CAD in ED patients with elevated risk for ACS, guiding subsequent patient management and avoiding unnecessary downstream cardiac tests.

This retrospective analysis of a coronary CTA registry included 2,290 ED patients who underwent CTA for ACS assessment from October 2012 through March 2017 at our tertiary hospital. We defined patients with reported history of CAD (medical records or self-reported), elevated initial troponin T (TnT) (>0.03 ng/ml) (Elecsys TnT fourth-generation, Roche, Indianapolis, Indiana), and/or abnormal electrocardiograms (ECGs) (ST-segment elevation ≥0.5 mm or T-wave inversion ≥1 mm) as intermediate to high risk; all others as low risk. Patients were initially assessed by emergency physicians (who were then instructed to consult a cardiologist before proceeding in these higher risk patients) (2). CTAs were performed on 128- or 192-slice dual-source computed tomography scanners (Somatom Definition Flash/Force, Siemens Healthineers, Forchheim, Germany) without restrictions to patient’s heart rate, rhythm, or body habitus, and read by cardiovascular-trained readers (2). Patients were classified by their highest degree of stenosis defining 0% to 49% luminal narrowing as nonobstructive and 50% to 100% as obstructive. Medical records were adjudicated for discharge diagnosis by independent cardiologists (2). The primary endpoints were diagnostic performance defined as sensitivity, specificity, positive and negative predictive values of CTA in intermediate-to-high-risk patients for the detection of: 1) ACS; and 2) of obstructive CAD undergoing revascularization within 30 days.

Overall, 124 of 2,290 (5.4%) patients (age 56.3 ± 11.1 years; n = 75 [60.5%] men; body mass index 28.6 ± 6.3 kg/m²) met criteria for intermediate to high risk as follows: ECG changes in 42 (33.9%), reported history of CAD in 80 (64.5%), and elevated initial TnT in 18 (14.5%) patients. Intermediate-to-high-risk patients were older (p = 0.003), had higher frequency of dyslipidemia and hypertension (both p < 0.001), higher thrombolysis in myocardial infarction risk scores (p < 0.001), and more coronary calcium by computed tomography scans (p = 0.007) compared with 2,166 (94.6%) patients classified as low risk. Coronary CTA detected obstructive CAD in 50 of 124 (40.3%) intermediate-to-high-risk patients and excluded obstructive stenosis in 74 of 124 (59.7%). Prevalence of obstructive CAD was 26.2% for abnormal ECG, 66.7% for elevated troponins, and 48.8% for reported history of CAD. Overall, 36 of 124 (29.0%) intermediate-to-high-risk patients had ACS, with 34 of 36 (94.4%) having obstructive CAD, whereas 2 of 36 (5.6%) had nonobstructive CAD but elevated TnT and, thus, diagnosed as type II myocardial infarct (secondary to ischemic imbalance). Subsequent cardiac testing was observed in 35 of 50 (70.0%) obstructive CAD patients, including 28 of 50 (56.0%) who had invasive coronary angiography (ICA), whereas only 2 of 70 (2.9%) nonobstructive CAD patients had additional noninvasive cardiac testing and none underwent ICA. ICA confirmed CTA-reported obstructive stenosis in 27 of 28 (96.4%) patients. Sixteen of 124 (12.1%) patients received revascularization within 30 days (Table 1). Sensitivity, specificity, positive predictive value, and negative predictive value of CTA for the detection of ACS was 94.4% (95% confidence interval [CI]: 81.3% to 99.3%), 81.8% (95% CI: 72.2% to 89.2%), 68.0% (95% CI: 57.5% to 76.9%), and 97.3% (95% CI: 90.3% to 99.3%), and for the detection of obstructive CAD undergoing revascularization within 30 days, 100.0% (95% CI: 79.4% to 100.0%), 68.5% (95% CI: 58.9% to 77.1%), 32.0% (95% CI: 26.3% to 38.3%), and 100.0% (95% CI: 100.0% to 100.0%), respectively.

This single-center experience revealed a good discriminatory capability of CTA to exclude obstructive CAD in approximately two-thirds of patients at intermediate to high risk, with a negative predictive value of 97% for ACS. Clinical factors and biomarkers were essential elements to safely exclude ACS and discharge. Although CTA is established for low-risk patients, CTA was used as an alternative strategy to standard care in intermediate-to-high-risk patients; therefore, we could expect that only a limited number of patients with this particular risk profile were triaged to CTA.

We speculate that CTA may serve as a “gate-keeper” to ICA in patients at higher risk for ACS or inconclusive results. Although this sample showed promising results of a coronary CTA strategy in intermediate-to-high-risk ED patients, it is important to recognize that forthcoming randomized controlled data are necessary to inform care standards (3).

• 2018 American College of Cardiology Foundation