High-risk Plaque Detected on Coronary Computed Tomography Angiography Predicts Acute Coronary Syndrome Independent of Significant Stenosis in Patients with Acute Chest Pain – Results from ROMICAT II Trial

Abstract

Background

To determine whether high-risk plaque as detected by coronary computed tomography angiography (CTA) permits improved early diagnosis of acute coronary syndrome (ACS) independent to the presence of significant CAD in acute chest pain patients.

Objectives

The primary aim was to determine whether high-risk plaque features, as detected by CTA in the emergency department, may improve diagnostic certainty of ACS independent and incremental to the presence of significant CAD and clinical risk assessment in patients with acute chest pain but without objective evidence of myocardial ischemia or myocardial infarction.

Methods

We included patients randomized to the CCTA arm of ROMICAT II trial. Readers assessed coronary CTA qualitatively for the presence of non-obstructive CAD (1-49% stenosis), significant CAD (≥50% or ≥70% stenosis), and the presence of at least 1 of the high-risk plaque features (positive remodeling, low < 30 Hounsfield Units plaque, napkin-ring sign, spotty calcium). In logistic regression analysis, we determined the association of high-risk plaque with ACS [myocardial infarction (MI) or unstable angina pectoris (UAP)] during the index hospitalization and whether this was independent of significant CAD and clinical risk assessment.

Results

Overall 37 of 472 patients who underwent coronary CTA with diagnostic image quality (mean age 53.9±8.0 years, 52.8% men) had ACS (7.8%; MI n=5, UAP n=32)]. CAD was present in 262 (55.5%) patients [non-obstructive CAD 217 (46.0%) patients, significant CAD with ≥50% stenosis 45 (9.5%) patients]. High-risk plaques were more frequent in patients with ACS and remained a significant predictor of ACS (OR 8.9, 95% CI 1.8-43.3, p=0.006) after adjusting for ≥50% stenosis (OR 38.6, 95% CI 14.2-104.7, p<0.001) and clinical risk assessment (age, gender, number of cardiovascular risk factors). Similar results were observed after adjusting for ≥70% stenosis.

Conclusions

In patients presenting to the ED with acute chest pain but negative initial electrocardiogram and troponin, presence of high-risk plaque on coronary CTA increases the likelihood of ACS independent of significant CAD and clinical risk assessment (age, gender, and number of cardiovascular risk factors).

Introduction

Coronary computed tomography angiography (CTA) is a viable alternative to functional testing with or without imaging in the evaluation of patients with acute chest pain in the emergency department (ED) (1-3). However, these trials revealed the uncertainty around management decisions in patients with significant CAD by CTA, as the positive predictive value of CTA for significant CAD remains moderate. Furthermore, the Rule Out Myocardial Infarction/Ischemia Using Computer Assisted Tomography I (ROMICAT I) trial demonstrated that the presence of stenosis >50% had limited diagnostic value for acute coronary syndrome (ACS) as only 46% of patients who had obstructive CAD by CTA had matching single-photon emission computed tomography (SPECT) perfusion abnormalities during stress testing (4). Therefore, the optimal management of patients with significant CAD on CTA in this setting remains uncertain. An opportunity for more efficient use of CTA in the management of patients with chest pain may arise from its capability to accurately assess plaque characteristics, such as positive remodeling, spotty calcium, low HU (Hounsfield Units) attenuation, and napkin-ring sign (5-11). Similar characteristics have been demonstrated to represent high-risk plaque on histology (necrotic lipid rich core, thin-cap fibroatheroma, positive remodeling, spotty calcium) (12,13) and intravascular imaging (positive remodeling, larger plaque volume, spotty calcium) (14,15). Furthermore, initial evidence from CTA studies suggests that these features were associated with an increased risk for future cardiovascular events in patients with stable chest pain syndromes (16-18) However, there are limited data on the potential use of high-risk plaque on CTA in patients with acute chest pain (19,20).

This study's primary aim was to determine whether high-risk plaque features, as detected by CTA in the ED, may improve diagnostic certainty of ACS independent and incremental to the presence of significant CAD and clinical risk assessment in patients with acute chest pain but without objective evidence of myocardial ischemia or myocardial infarction (MI).

Methods

The study cohort consisted of patients who were randomized to the CTA arm of the ROMICAT II trial and underwent CTA (Figure 1). A detailed description of the patient population was reported (2). Between April 2010 and January 2012, 1,000 patients with cardiovascular risk factors presenting to the ED of 9 U.S. hospitals with chest pain and a clinical suspicion for ACS were enrolled. All study participants provided written consent for participation in ROMICAT II. The local institutional review boards approved the study.

Figure 1.

Study population enrollment, exclusion and inclusion.

CTA images were acquired using either retrospectively ECG-gated or prospectively ECG-triggered protocols. The investigators in the study used the scanners from 3 vendors (Siemens Healthcare, Erlangen, Germany; GE Healthcare, Waukesha, WI Toshiba America Medical Systems, Tustin, CA) and different scanner generations (64-, 128-, 256-row, and dual-source). The images were transferred to the core lab. Image analysis was performed on a cardiac workstation (TeraRecon, Foster City, CA). Three readers with at least 5 years of experience and level III training in CTA analyzed the datasets. Each reader analyzed one-third of randomly assigned CTA datasets. Further, 30 randomly selected CTA datasets were analyzed by all 3 readers to determine interobserver agreement. The CTA analysis was performed per coronary segment using the model of the Society of Cardiovascular Computed Tomography (21). For each coronary segment, the reader determined whether the image quality was sufficient to evaluate for the presence of stenosis and coronary plaque with confidence. Coronary segments that were assessed as non-diagnostic in image quality were treated as non-informative for the purpose of the analysis.

Each evaluable coronary segment was assessed for the presence of stenosis. The severity of stenosis was quantified by visual estimation and divided into 4 categories: 0%=no stenosis, 1-49%=mild stenosis, 50-70%=moderate stenosis, and ≥70%=severe stenosis/occlusion. For each evaluable coronary segment, we noted the presence of plaque. Non-calcified coronary plaque was defined as any discernible structure that could be assigned to the coronary artery wall, had a CT number below the contrast-enhanced coronary lumen but above the surrounding connective tissue, and could be identified in at least 2 independent planes (22). Any structure with a density of ≥130 HU that could be visualized separately from the contrast-enhanced coronary lumen, could be assigned to the coronary artery wall, and could be identified in at least 2 independent planes was defined as calcified atherosclerotic plaque (22). In each coronary segment with plaque, we performed further qualitative evaluation for the presence of high-risk plaque features, which were defined as positive remodeling, low CT number of plaque, napkin-ring sign, and spotty calcium (Central Illustration). Positive remodeling was assessed visually in multi-planar reformatted images reconstructed in long axis and short axis view of the vessel. Additional manual measurements of outer vessel diameter were performed at readers' discretion, and threshold of 1.1 was used to define positive remodeling.(5,6) If low CT attenuation was visually noted in non-calcified plaque, readers placed 3 random region-of-interest measurements (approximately 0.5-1.0 mm²) in the non-calcified low CT attenuation portion of the plaque. Low HU plaque was defined as the mean CT number within these 3 regions of interest <30 HU (7,16). The napkin ring sign was defined as a ring-like peripheral higher attenuation of the non-calcified portion of the coronary plaque (20,23-25). Spotty calcium was defined as the presence of calcified plaque with a diameter <3 mm in any direction, length (extent in the longitudinal direction of the vessel) of the calcium less than 1.5 times the vessel diameter and width (extent of the calcification perpendicular to the longitudinal direction of the vessel) of the calcification <2/3 of the vessel diameter (8,9,26). The patient was classified as having high-risk plaque features, if at least 1 high-risk plaque feature was present.

The primary outcome of the study was an ACS event during the index hospitalization. ACS was defined as acute MI or unstable angina pectoris according to the American College of Cardiology/American Heart Association Guidelines (2,27). An independent clinical-events committee predefined and adjudicated the endpoint. We excluded ACS (myocardial infarction) during the index hospitalization in a patient with an anomalous right coronary artery from the main pulmonary artery, but with no evidence of coronary plaque or stenosis, who underwent the re-implantation of the right coronary artery during the index hospitalization.

All statistical analyses were performed using Stata 13.1 (StataCorp LP, College Station, TX). Continuous data are presented as mean ± standard deviation. Comparisons between groups were performed with the use of an independent sample t-test for continuous variables, Fisher's exact test for categorical variables, and the Wilcoxon rank-sum test for ordinal variables. To determine whether the presence of high-risk plaque is an independent predictor of ACS, we performed multivariable logistic regression analyses and adjusted for the presence of ≥50% stenosis, age, gender, and the number of cardiovascular risk factors (hypertension, dyslipidemia, diabetes mellitus, smoking status, and family history of premature CAD). We also performed the analyses with a definition of the significant CAD as ≥70% stenosis or left main coronary stenosis ≥50%. To determine whether high-risk plaque is incremental to presence of significant CAD and clinical risk assessment, we compared areas under the receiver operating characteristics curve (AUC) using the DeLong algorithm (31). For all analyses, a 2-tailed p value <0.05 was required to reject the null hypothesis.

Results

From 501 patients randomized to CTA, 473 underwent CTA. Reasons for not undergoing a CTA were: patient declined CTA (n=9), safety concerns (n=5), unavailability of CTA (n=5), or technical difficulties (n=9). One patient was excluded from further analysis due to non-diagnostic image quality in all coronary segments. CTA was performed using the scanners from 3 vendors (Siemens 58%, GE 34%, Toshiba 7%). Overall, 472 patients who underwent CTA with diagnostic image quality formed the study population (mean age 53.9±8.0 years, 52.8% men). The prevalence of ACS was 7.8% (n=37; MI n=5; unstable angina pectoris n=32). Patients with ACS had more cardiovascular risk factors and higher thrombolysis in myocardial infarction (TIMI) scores (Table 1).

Table 1. Clinical characteristics of study patients stratified according to the diagnosis of acute coronary syndrome.

	ACS (n=37)	No ACS (n=435)	p value
Age (years)	57.2 ± 8.3	53.6 ± 7.9	0.015
Women, n (%)	6 (16.2)	217 (49.9)	<0.001
Cardiovascular risk factors, n (%)
Hypertension	22 (59.5)	230 (52.9)	0.50
Diabetes mellitus	7 (18.9)	72 (16.6)	0.65
Dyslipidemia	25 (67.6)	191 (43.9)	0.006
Former or current smoker	25 (67.6)	211 (48.5)	0.039
Family history of premature CAD	11 (29.7)	120 (27.6)	0.85
Number of cardiovascular risk factors, %			0.005
0 or 1	21.6	37.5
2 or 3	62.2	53.6
≥ 4	16.2	9.0
TIMI score, %			<0.001
0	35.1	63.9
1	46.0	27.1
2	18.9	8.1
3	0.0	0.9
Troponin classification, n (%)			<0.001
Negative	28 (75.7)	426 (97.9)
Borderline	6 (16.2)	8 (1.9)
Elevated	3 (8.1)	1 (0.2)
Invasive coronary angiography, n (%)	32 (86.5)	19 (4.4)	<0.001
Significant CAD (≥ 50% stenosis) in angiography	31 (96.9)	5 (26.3)	<0.001
Percutaneous coronary intervention, n (%)	22 (59.5)	0 (0.0)	<0.001
Coronary artery by-pass graft surgery, n (%)	4 (10.8)	1 (0.2)	<0.001

ACS = Acute Coronary Syndrome; CAD = coronary artery disease; TIMI = Thrombolysis in Myocardial Infarction

Overall 202 of 6,855 (2.9%) coronary segments were non-evaluable with at least 1 non-evaluable segment present in 71 of 472 patients (15.0%). The most common reason for non-evaluability was the presence of motion artifacts (n=153) followed by coronary calcium (n=31) and poor contrast-to-noise ratio (n=17). Interobserver agreement among 3 readers in 30 patients was very good for the presence of coronary plaque (κ=0.77) and significant CAD (stenosis ≥50%) (κ=0.80) and good for high-risk plaque (κ=0.69).

CAD, defined as the presence of any plaque, was detected in 262 (55.5%) patients. Among 262 patients with coronary plaque, calcified plaques were present in 217 (82.2%) and non-calcified plaques in 199 (76.0%) patients. Non-obstructive CAD (1-49% stenosis) was detected in 217 of 472 (46.0%) patients. Significant CAD (≥50% stenosis) was found in 45 of 472 (9.5%) patients. Stenosis of ≥70% or left main coronary stenosis of ≥50% was detected in 24 (5.1%) patients. At least 1 high-risk plaque feature was present in 167 of 472 (35.4%) patients, at least 2 high-risk plaque features were present in 56 (11.9%) patients, and at least 3 high-risk plaque features were present in 35 (7.4%) patients. The most common feature was spotty calcium (n=151; 32.0%) followed by positive remodeling (n=55, 11.7%), low HU plaque (n=40, 8.5%), and napkin ring sign (n=26, 5.5%).

Among 45 patients with significant CAD (≥50% stenosis), at least 1, 2, and 3 or more high-risk plaque features were present in 41 (91.1%), 27 (60.0%), and 24 (53.3%) patients, respectively. The prevalence of the individual high-risk plaque features in patients with significant CAD was as follows: spotty calcium in 40 (88.9%), positive remodeling in 28 (62.2%), low HU plaque in 19 (42.2%), and napkin ring sign in 16 (35.6%) patients. We observed a lower prevalence of high-risk plaque features among 217 subjects with non-obstructive CAD (1-49% stenosis). At least 1, 2, and 3 high-risk plaque features were present in 125 (57.6%), 29 (13.4%), and 11 (5.1%) patients, respectively. The prevalence of the individual high-risk plaque features in patients with non-obstructive CAD was as follows: spotty calcium in 111 (51.2%), positive remodeling in 27 (12.4%), low HU plaque in 21 (9.7%), and napkin ring sign in 10 (4.6%) patients.

A comparison of the CTA findings in patients with and without ACS is provided in Table 2. All patients with ACS had evidence of CAD (i.e., either coronary plaque with 1-49% stenosis or ≥50% stenosis). In the non-ACS group, almost half of the patients had no evidence of CAD (p<0.001). More than 75% of ACS patients had significant CAD with ≥50% stenosis as compared to <4% of patients without ACS (p<0.001).

Table 2. Coronary computed tomography angiography characteristics of patients stratified according to the diagnosis of acute coronary syndrome.

	ACS (n=37)	No ACS (n=435)	p value
Coronary artery disease (CAD) category, n (%)
No CAD	0 (0.0)	210 (48.3)	<0.001
Non-obstructive CAD (1-49% stenosis)	8 (21.6)	209 (48.1)	0.002
Significant CAD (≥ 50% stenosis)	29 (78.4)	16 (3.7)	<0.001
Plaque category, n (%)
No plaque	0 (0.0)	210 (48.3)	<0.001
Calcified plaque	36 (97.3)	181 (41.6)	<0.001
Non-calcified plaque	36 (97.3)	163 (37.5)	<0.001
High-risk plaque category, n (%)
Any high-risk plaque	35 (94.6)	132 (30.3)	<0.001
Napkin-ring sign	12 (32.4)	14 (3.2)	<0.001
Positive remodeling	22 (59.5)	33 (7.6)	<0.001
Low HU plaque	16 (43.2)	24 (5.5)	<0.001
Spotty calcium	35 (94.6)	116 (26.7)	<0.001

ACS = Acute coronary syndrome; CAD = Coronary artery disease; HU = Hounsfield units

At least 1 high-risk plaque feature was present in 95% of patients with ACS and in 30% of patients without ACS (p<0.001). All individual high-risk plaque features were more often observed in ACS patients (p<0.001). Furthermore, all patients with ACS and without ≥50% stenosis had at least 1 high-risk plaque.

The presence of ACS was strongly associated with the presence of a significant CAD (≥50% stenosis) as detected by CCTA. In univariate analysis, patients with a significant CAD (≥50% stenosis) were 34 times more likely to have an ACS during the index hospitalization as compared to those without a significant CAD (Figure 2). Similar to significant CAD, the presence of any high-risk plaque was associated with ACS. Patients with at least 1 high-risk plaque feature were 32 times more likely to have an ACS during the index hospitalization. All individual high-risk plaque features were associated with ACS.

Figure 2.

Probability of having acute coronary syndrome during the index hospitalization according to coronary computed tomography characteristics.

Central Illustration: Significant stenosis and high-risk coronary plaque features and their association with probability of having acute coronary syndrome during the index hospitalization.

Stenosis ≥50% – Severe stenosis of the mid left anterior descending coronary artery (red arrow). Non-calcified plaque with positive remodeling in the distal right coronary artery (arrowhead). Positive remodeling – The two dotted red lines (image insert) demonstrate the vessel diameters at the proximal and distal reference (both 1.8 mm) and the full red line demonstrates the maximal vessel diameter in the mid portion of the plaque (2.7 mm) – the remodeling index is 1.5 Low HU plaque – Partially calcified plaque in the mid right coronary artery with low <30 HU plaque. The red circles demonstrate the three regions-of-interest with the mean CT number of 22 HU, 19 HU, and 20 HU

Napkin ring sign – Napkin ring sign plaque in the mid left anterior descending coronary artery. Schematic cross-sectional view of the napkin ring sign. The red line demonstrates the central low HU area of the plaque adjacent to the lumen (yellow ellipse) surrounded by a peripheral rim of the higher CT attenuation (red arrows).

Spotty calcium – Partially calcified plaque in the mid right coronary artery with spotty calcification (diameter <3 mm in all directions; red circles).

In the logistic regression analysis (Table 3), the presence of high-risk plaque (odds ratio 8.9, 95% CI 1.8 to 43.3, p=0.006) remained significantly associated with ACS after adjusting for a ≥50% stenosis (odds ratio 38.6, 95% CI 14.2 to 104.7, p<0.001) and clinical predictors (age: odds ratio 1.0, 95% CI 0.9 to 1.1, p=0.87, female gender: odds ratio 0.4, 95% CI 0.1 to 1.2, p=0.104, number of cardiovascular risk factors: odds ratio 1.3, 95% CI 0.8 to 2.0, p=0.278). Similar results were observed when significant CAD was defined as a stenosis ≥70% or left main coronary stenosis ≥50%.

Table 3.

Multivariable logistic regression analysis for the prediction of ACS using clinical predictors in model 1 (age, gender, and number of cardiovascular risk factors that include diabetes mellitus, hypertension, dyslipidemia, smoking status, and family history of premature CAD), adding ≥50% stenosis in model 2 and high-risk plaque in model 3.

	Model 1		Model 2		Model 3
	OR (95% CI)	p value	OR (95% CI)	p value	OR (95% CI)	p value
Age	1.1 (1.0-1.1)	0.003	1.0 (1.0-1.1)	0.539	1.0 (0.9-1.1)	0.870
Female	0.2 (0.1-0.4)	<0.001	0.3 (0.1-0.8)	0.020	0.4 (0.1-1.2)	0.104
Number of Risk Factors	1.4 (1.0-1.8)	0.056	1.4 (0.9-2.2)	0.124	1.3 (0.8-2.0)	0.278
Stenosis ≥50%			71.7 (27.1-189.9)	<0.001	38.6 (14.2-104.7)	<0.001
High-risk plaque					8.9 (1.8-43.3)	0.006

ACS = acute coronary syndrome, CI = confidence interval, Number of Risk Factors = number of cardiovascular risk factors (diabetes mellitus, hypertension, dyslipidemia, smoking status, and family history of premature CAD), OR = odds ratio,

We demonstrated that coronary stenosis ≥50% was incremental to baseline clinical characteristics (age, gender, number of cardiovascular risk factors) in predicting ACS (model 2: AUC 0.935, 95% CI 0.894 to 0.976 vs. model 1: AUC 0.776, 95% CI 0.711 to 0.840, p<0.001). We observed that adding the presence of high-risk plaque to the model further improved the prediction of ACS (model 3: AUC 0.959, 95% CI 0.937 to 0.981 versus model 1, p<0.001, model 3 versus model 2, p=0.03). Similar results were observed when significant CAD was defined as a stenosis ≥70% or left main coronary stenosis ≥50%.

Discussion

We demonstrated that high-risk coronary plaque as detected on CTA in patients presenting to the ED with acute chest pain was associated with ACS independently and incrementally to the presence of significant CAD and clinical risk assessment. Our results suggested that CTA based assessment of high-risk plaque improved diagnosis of ACS in patients with acute chest pain who otherwise have no electrocardiographic or enzymatic evidence of ischemia or infarction. Our understanding of morphologic features of high-risk plaque stems primarily from the histology studies of patients who died from sudden cardiac death. The histologic features of the culprit plaques included large necrotic core, higher macrophage count, positive remodeling, speckled calcium, and thin fibrous cap (12,13). Similar morphologic features (positive remodeling, larger plaque area, spotty calcium, and large necrotic core) were observed with intravascular imaging in culprit lesions of ACS (14,15).

The direct comparison of CTA features of high-risk plaque to virtual histology intravascular ultrasound (VH-IVUS) is challenging. In the PROSPECT trial, the authors demonstrated that lesions associated with recurrent ACS were characterized by a plaque burden of ≥70%, a minimal luminal area of ≤4.0 mm² or to be classified as thin-cap fibroatheroma (29). CTA characteristics of high-risk plaque in our study did not exactly correlate to VH-IVUS measurements. However, the presence of a minimal luminal area of ≤4.0 mm² often correlates with the significant stenosis and we demonstrated that significant stenosis was an independent predictor of ACS. We did not perform quantitative analysis of plaques, which is necessary for the calculation of plaque burden. However, there is a correlation between positive remodeling and large plaque burden. Finally, the spatial resolution of CTA does not permit the detection of thin-cap fibroatheroma. However, the presence of the napkin-ring sign was very specific for the presence of advanced atheroma (23).

High-risk plaque features have been the target of non-invasive imaging with CTA. An early CTA study showed the feasibility of detecting high-risk plaque (5). The culprit plaques of ACS were often positively remodeled and had larger plaque burden as compared to similarly stenotic plaques in patients with stable angina. Subsequent studies extended this observation and showed an association of plaque features such as large plaque burden, positive remodeling, spotty calcium, low HU plaque and napkin ring sign with ACS (7,9,19,20) and also with an increased risk of future cardiovascular events (18-20). Limited data exist on the role of high-risk plaque for early diagnosis of ACS in the acute chest pain population (19,20).

We found that one-third of the acute chest pain patients had high-risk plaque, with spotty calcium as the most frequent high-risk plaque feature (32.0%), followed by positive remodeling (11.7%), low HU plaque (8.5%), and napkin ring sign (5.5%). The prevalence of high-risk plaque features observed in our study in the acute chest pain population is similar to other CTA studies (5-15%) performed in populations of patients undergoing invasive coronary angiography, larger unselected patients populations, and in non-culprit vessels of patients with ACS (PROSPECT trial) (8,9,16,20,29). The reported prevalence of spotty calcium varied more dramatically in the literature (0-43%), most likely as a result of differences in the definition of spotty calcium (9,20). In addition, we observed a prevalence of coronary atherosclerosis defined as any coronary plaque in 55.5% of patients, non-obstructive CAD in 46.0% of patients and significant stenosis ≥50% in 9.5% of patients, mostly consistent with other single and multicenter trials in acute chest pain (1,3,30). Overall, our results demonstrate that the prevalence of high-risk plaque and its associations with CAD and ACS in the acute chest pain setting are generalizable to multiple centers and CT vendors, and they are also in accordance with observations in stable chest pain syndromes as well as in non-culprit vessels of patients with ACS undergoing percutaneous coronary interventions.

The assessment of patients presenting with acute chest pain, but without objective signs of ischemia or MI, remains a diagnostic challenge. In patients who present with acute chest pain, exclusion of a significant coronary stenosis and plaque by CTA has a high sensitivity and negative predictive value for ACS and allows early discharge (1-3). However, ACS cannot be ruled out in a significant portion of patients, in whom coronary plaques are present, reducing the specificity of CTA. A significant coronary stenosis was detected in approximately 10% and 4% of patients in 2 large multicenter trials (1,3). However, ACS or major adverse cardiovascular events developed only in approximately 4% and 1% of patients. Conversely, the sensitivity of ≥50% stenosis for the detection of ACS was 77% in the ROMICAT I trial (33). In ROMICAT II, we found a very similar result with a ≥50% stenosis detected in 78% of patients with ACS. This finding concurs with previous invasive angiographic studies that observe an absence of a significant stenosis in 12-14% ACS patients (31,32). The limited diagnostic accuracy of CTA using the traditional criterion of a significant stenosis might increase downstream testing and interventions (1-3,33). A potential means to improve CTA diagnostic accuracy is by adding the assessment of high-risk plaque to stenosis, which improved the diagnostic assessment of patients with acute chest pain in the present study.

The value of high-risk plaque for the diagnosis of ACS in patients with a significant stenosis was demonstrated in the ROMICAT I trial (21). A score including positive remodeling, spotty calcium, volume of low HU plaque, and stenosis length had a good discriminatory capacity for ACS during index hospitalization, but was only limited to those with a significant stenosis on CTA. The addition of high-risk plaque features showed a potential to refine the diagnosis of ACS by CTA (19,20). The current study demonstrates that high-risk plaque features assessed by a qualitative read of CTA images were independent and incremental to significant stenosis and clinical risk assessment for predicting ACS during the index hospitalization. While stenosis remained the strongest predictor of ACS, high-risk plaque was associated with 9-fold increase in the likelihood of ACS after adjusting for the presence of stenosis ≥50% and clinical risk assessment.

The inclusion of high-risk plaque improved the detection of ACS in patients with mild stenosis (1-49%). All patients with mild stenosis and ACS had at least 1 high-risk plaque feature. We suggest that patients with mild stenosis and high-risk plaque cannot be safely discharged from the ED. Further evaluation with serial troponins and additional testing will be necessary. While awaiting further work-up, providers should consider aggressive medical therapy (e.g., dual antiplatelet therapy). On the other side of the spectrum, there are patients with significant stenosis. Patients with a significant stenosis on CTA cannot be discharged from the ED after the initial troponin and electrocardiogram (1-3). In our study, the presence of high-risk plaque was incremental to stenosis for the prediction of ACS. Therefore, providers should consider aggressive medical therapy and invasive coronary angiography in patients with significant stenosis and high-risk plaque. In patients with stenosis, but no high-risk plaque, providers should consider further work-up to confirm the significance of the stenosis (e.g., stress test or invasive coronary angiography with fractional flow reserve). However, these strategies have not been tested in the prospectively designed clinical studies and further studies are required to include them in routine clinical practice.

Limitations

The low number of ACS outcomes (n=37) limited our ability to perform sub-analyses and to include additional variables in the multivariable models. Recent studies demonstrated the additional value of quantitative analysis of plaque by CTA (5,6,20-23), we performed qualitative assessment of stenosis and high-risk plaque. The decision to use qualitative assessment was motivated by the fact that this approach could be more feasible in routine clinical practice as it adds minimal time for the assessment and does not require specific software and hardware. We restricted our analysis to the 4 most established high-risk plaque features (positive remodeling, low HU plaque, napkin ring sign, and spotty calcium).

Conclusions

The presence of high-risk plaque on CTA increases the likelihood of ACS independent and incremental to the presence of significant CAD and clinical risk assessment (age, gender, number of cardiovascular risk factors) in patients with acute chest pain and with no objective evidence of myocardial ischemia or infarction.

Perspectives

Competency in Medical Knowledge

Among patients presenting with acute chest pain, normal initial cardiac troponin levels and no evidence of ischemia on the electrocardiogram, coronary computed tomography angiography (CTA) can identify high-risk coronary atherosclerotic lesions independently associated with acute coronary syndromes and diagnostic information incremental to the detection of ≥50% luminal stenosis.

Translational Outlook

Multicenter randomized trials are needed to determine whether medical therapy and/or interventions based on CTA characterization of high-risk plaque improves clinical outcomes in patients with acute chest pain.

Abbreviations

CTA

computed tomography angiography

ACS

acute coronary syndrome

CAD

coronary artery disease

HU

Hounsfield units

ROMICAT

Rule Out Myocardial Infarction/Ischemia Using Computer Assisted Tomography

CI

confidence interval

VH-IVUS

virtual histology – intravascular ultrasound

ED

emergency department