Early versus late acute coronary syndrome risk patterns of coronary atherosclerotic plaque 

Abstract

Graphical Abstract

Early versus late ACS risk patterns of coronary atherosclerotic plaque. Schematic representation of the study design and patients (left panel), specifics of the two cohorts (middle panel), and the volumetric differences in composition of coronary atherosclerosis between patients with early versus late ACS (right panel). ACS, acute coronary syndrome; CCTA, coronary computed tomography angiography; MACE, major adverse cardiac events.

Open in new tabDownload slide

See the editorial comment for this article ‘Coronary plaque in the fourth dimension: associating coronary computed tomography plaque components and the time to acute coronary syndrome’, by Jay Voit and Kelley R.H. Branch, https://doi.org/10.1093/ehjci/jeac170.

Introduction

It has been believed for decades that the majority of acute coronary syndromes (ACS) results from rupture of coronary lesions with a previously documented mild degree of luminal stenosis.¹ However, discrepancies exist amongst studies with regard to their timing antecedent to the ACS.^2–4 To this end, the volumetric progression of coronary lesions is thought to be the fundamental step for whether or not an ACS will occur.^5–7 Coronary computed tomography angiography (CCTA) has the ability to non-invasively detect, characterize and quantify coronary atherosclerotic plaque.^8,9 In this context, a larger volume of non-calcified plaque on CCTA has been related to vulnerable plaque and future myocardial infarction (MI).^10–13 We previously reported findings from the Incident COroNary Syndromes Identified by Computed Tomography (ICONIC) study, highlighting the importance of a comprehensive coronary plaque evaluation for the identification of patients at high risk for ACS.¹² The current analysis sought to examine differences in the volume and composition of coronary atherosclerosis between patients experiencing an early versus late ACS after clinically indicated baseline CCTA.

Methods

Study design and patients

Derivation cohort, ICONIC study

ICONIC is a nested case-control study within the prospective, dynamic, multicenter, observational COronary CT Angiography EvaluatioN For Clinical Outcomes: an InteRnational Multicenter registry, enrolling patients at 13 sites in 8 countries across Asia, Europe, and North-America between 2002 and 2009.^12,14 The study design including in- and exclusion criteria has been published in detail.¹² Of those enrolled in the ICONIC study, a total of 234 patients underwent a clinically indicated baseline CCTA for suspected coronary artery disease (CAD) and experienced an ACS during follow-up. Separate core laboratories performed blinded adjudication of ACS events and culprit lesions according to World Health Organization (WHO) and Third Universal MI definitions, as well as blinded qualitative and quantitative analysis of baseline CCTA according to an 18-segment Society of Cardiovascular Computed Tomography (SCCT) model.^15–17 ACS patients were 1:1 propensity-matched to within-site non-ACS controls based on age, sex, cardiac risk factors and CAD severity on CCTA defined as non-obstructive <50%, 1-vessel, 2-vessel, and 3-vessel or left main obstructive disease ≥50%. The study protocol was approved by institutional review boards or ethics committees at each participating site, and all patients provided written informed consent. For the present analysis, non-ACS controls (n = 234) were omitted. Thus, 234 ACS patients were included. Data including the non-ACS controls are presented in Supplementary data online, Table S1.

External validation cohort, PARADIGM registry

The cohort used to externally validate differences in the volume and composition of coronary atherosclerosis between patients experiencing early versus late ACS was the Progression of AtheRosclerotic PlAque DetermIned by Computed TomoGraphic Angiography Imaging (PARADIGM) registry. PARADIGM is a prospective, dynamic, multicenter, observational registry, enrolling patients for serial CCTA with an interscan interval of ≥2 years at 13 sites in 7 countries across Asia, Europe, North- and South-America between 2003 and 2015.¹⁸ The study design including in- and exclusion criteria has been published in detail.¹⁹ Of those enrolled in the PARADIGM registry, a total of 452 patients underwent a clinically indicated baseline CCTA for suspected CAD and experienced a major adverse cardiac event (MACE) during follow-up. MACE included MI, cardiac, and non-cardiac death, as well as early and late percutaneous coronary intervention or coronary artery bypass grafting surgery. The study protocol was approved by each site’s institutional review board or ethics committee, and all patients provided written informed consent. For the present analysis, patients with a non-interpretable baseline CCTA due to prior stents or grafts (n = 35), artefacts (n = 28), missing or file-errors of images (n = 12) or a combination of the reasons above (n = 13) were excluded. Thus, 364 MACE patients were included.

Event adjudication

For the derivation cohort, a detailed methodology for the adjudication of ACS events and culprit lesions was previously reported.¹² First, ACS events were adjudicated by six experienced physicians at the Clinical and Data Coordinating Center (Dalio Institute of Cardiovascular Imaging, New York, NY, USA) according to WHO and Third Universal MI definitions.^15,17 Thorough review of electrocardiograms, cardiac biomarkers, and invasive coronary angiograms (ICAs) was completed, blinded to baseline CCTA. Second, for each ACS patient 1 culprit lesion was adjudicated based on availability of the ICA. In case 1 significant lesion was observed on ICA, this was deemed the culprit lesion. In case ≥2 lesions were observed on ICA, the culprit lesion was deemed based on a combination of both stenosis severity and ischaemia localization on the electrocardiogram. Reasons for unfeasible adjudication of culprit lesions on ICA (n = 72) were earlier described.¹² Third, culprit lesions were co-registered to culprit lesion precursors on baseline CCTA using fiduciary landmarks such as the distance from the ostium and side branches. Reasons for unfeasible co-registration to culprit lesion precursors on baseline CCTA (n = 38) were earlier described.¹² For a total of 124 culprit lesions, successful co-registration was achieved.

For the external validation cohort, definitions of MACE events and methodology for follow-up were previously reported.¹⁹

CCTA acquisition and image analysis

Patients from both cohorts were scanned with ≥64 detector row CT scanners in agreement with the SCCT guidelines and site-specific standards.^16,20 Scans were qualitatively and quantitatively analyzed by level III-experienced readers at the CCTA core laboratory (Severance Cardiovascular Hospital, Seoul, South-Korea) according to the 18-segment SSCT model, blinded to clinical data including event status.¹⁶ For quantitative analysis, semi-automated validated software with appropriate manual correction was used at every 0.5–1.0 mm cross-section (QAngio CT Research Edition version 2.1.9.1, Medis Medical Imaging Systems, Leiden, the Netherlands).⁸ All coronary segments ≥2 mm in diameter were evaluated, and a coronary lesion was defined as any tissue ≥1 mm² within or adjacent to the lumen that could be discriminated in >2 planes from the pericardial tissue, epicardial fat, or lumen.¹¹ Comprehensive evaluation included measurements of maximal diameter stenosis, maximal cross-sectional plaque burden, lesion length, total and compositional plaque volumes, and adverse plaque characteristics, amongst others. Maximal diameter stenosis and cross-sectional plaque burden were calculated in line with standardized definitions.³ Compositional plaque volumes were uniformly categorized by Hounsfield units (HU): calcified plaque >350 HU, fibrous plaque 131–350 HU, fibrofatty plaque 31–130 HU and necrotic core <30 HU.¹² Adverse plaque characteristics included low-attenuation plaque <30 HU, positive remodeling ≥1.1, and spotty calcification ≤3 mm in any direction. High-risk plaque was defined as any coronary lesion exhibiting ≥2 of the features above.¹¹ All measurements were performed on a per-lesion and per-segment level, and summation of these values from the whole coronary artery tree generated patient-level data. Excellent intra- and interobserver intraclass correlations for measurements were previously reported for both cohorts.^12,18

Study aims

The definition of early or late ACS was determined based on the time between the baseline CCTA and subsequent ACS. The cut-off for late ACS was fixed at 90 days after baseline CCTA in line with prior published reports, describing the interplay between the vulnerability and progression of coronary atherosclerosis.^5,6 Primary aim was the identification of differences in total and compositional plaque volumes between patients experiencing early (≤90 days) versus late ACS (>90 days). Secondary aims included identification of differences between proximal culprit lesion precursors of early and late ACS, within-patient identification of future culprit lesions and validation of (primary) patient-level findings using the external validation cohort.

Statistical analysis

Continuous data are reported as means ± standard deviations (SD) or medians with interquartile ranges (IQR). Categorical data are reported as counts with percentages. Continuous data were compared with the Independent-Samples T test or Mann–Whitney U test based on skewness of distribution. Categorical data were compared with the χ² test or Fisher’s Exact test based on minimum expected counts. First, patient-level comparisons were performed amongst patients with early and late ACS. Second, lesion-level comparisons were performed amongst all culprit lesion precursors of early and late ACS. Due to variability of absolute plaque volumes according to the location within the coronary artery tree, specific focus was given to proximal precursors in the left main artery, proximal left anterior descending artery, proximal right coronary artery, and proximal left circumflex artery (Supplementary data online, Figure S1). For visual interpretation of the time-dependent relationship, estimated plaque volumes derived from a general linear model were plotted against the time between the baseline CCTA and subsequent ACS.²¹ Third, generalized estimating equations with a first-order autoregressive correlation structure were calculated to identify the compositions associated with culprit lesion precursors of either early or late ACS using within-patient non-culprit lesions as a comparator. Last, an external validation of statistically significant patient-level compositions was performed in the PARADIGM registry using early (≤90 days) versus late MACE (>90 days) after baseline CCTA. All statistical tests were two-sided and a P-value of <0.05 indicated statistical significance. All analyses were performed with R (version 3.6.1, R Development Core Team, Vienna, Austria) and SPSS software (version 26, SPSS IBM Corp., Armonk, NY, USA).

Results

Patients

A total of 234 patients (mean age 62 ± 12 years, 36% women) underwent baseline CCTA and experienced an ACS after a mean of 395 ± 622 days. ACS comprised 40 (17%) ST-segment elevation myocardial infarctions (STEMI), 114 (49%) non-ST-segment elevation myocardial infarctions (NSTEMI), 6 (3%) unclassified MI due to timing of the electrocardiogram preceding the ACS, and 74 (32%) unstable angina pectoris events. Patients who had an early ACS were largely comparable to patients who had a late ACS, as the majority was symptomatic with prevalent cardiac risk factors (Table 1). Distribution of ACS type and infarction size measured by cardiac biomarkers was similar amongst groups, whilst the left anterior descending artery was more frequently the culprit vessel in patients with early ACS (Table 2).

Patient-level comparison of early versus late ACS on baseline CCTA

As compared to patients with late ACS, early ACS patients demonstrated more often obstructive disease ≥50% with a corresponding greater maximal diameter stenosis and maximal cross-sectional plaque burden (Table 3). Total, fibrous, fibrofatty and necrotic core volumes were significantly larger in the early ACS group [256.9 mm³ (IQR 86.2–398.4 mm³) vs. 138.3 mm³ (IQR 53.3–354.3 mm³), P = 0.026; 118.9 mm³ (IQR 36.0–192.4 mm³) vs. 66.0 mm³ (IQR 22.4–155.5 mm³), P = 0.031; 35.5 mm³ (IQR 9.0–90.6 mm³) vs. 21.9 mm³ (IQR 3.6–52.3 mm³), P = 0.011; and 1.9 mm³ (IQR 0.1–10.1 mm³) vs. 0.8 mm³ (IQR 0.0–3.6 mm³), P = 0.007, respectively) (Figure 1A). Similar findings were observed with regard to fibrofatty plaque and necrotic core when compositional plaque volumes were vessel volume-normalized (Figure 1B). The time-dependent relationship for total and non-calcified plaque volumes, using estimated values derived from a general linear model, is illustrated in Figure 2.

Figure 1

(A) Patient-level comparison of plaque volumes according to early versus late ACS. Patient-level comparison of plaque volumes between patients with early (blue bars, n = 129) versus late ACS (grey bars, n = 105). (B) Patient-level comparison of percent atheroma volumes according to early versus late ACS. Patient-level comparison of percent atheroma volumes between patients with early (blue bars, n = 129) versus late ACS (grey bars, n = 105). ACS, acute coronary syndrome.

Open in new tabDownload slide

Figure 2

Patient-level visualization of plaque volumes according to ACS timing. Scatterplot depicts estimated plaque volumes (y-axis; increased values = increased amount of coronary plaque in mm³) by the time between the baseline CCTA and subsequent ACS (x-axis; increased values = later occurrence of event in days). All data points are patient-level values, and estimated values were derived from a general linear model. Black, total plaque; dark-green, fibrous plaque; light-green, fibrofatty plaque; red, necrotic core. ACS, acute coronary syndrome; CCTA, coronary computed tomography angiography.

Open in new tabDownload slide

Lesion-level comparison of early versus late ACS on baseline CCTA

As compared to precursors of late ACS, early culprit lesion precursors exhibited a greater maximal diameter stenosis and maximal cross-sectional plaque burden (Supplementary data online, Table S2). Total and compositional plaque volumes were not significantly larger in the early ACS group (P > 0.05). However, focused on proximal culprit lesion precursors only, fibrofatty volumes were significantly larger in early ACS as compared to late ACS [14.6 mm³ (IQR 3.5–62.5 mm³) vs. 5.5 mm³ (IQR 1.0–34.4 mm³), P = 0.048] (Figure 3). A trend towards larger necrotic core volumes within the early ACS group was observed, that reached statistical significance when the volume of necrotic core was combined with fibrofatty plaque [0.5 mm³ (IQR 0.0–4.1 mm³) vs. 0.1 mm³ (IQR 0.0–1.7 mm³), P = 0.065; and 15.5 mm³ (IQR 3.5–66.7 mm³) vs. 5.9 mm³ (IQR 1.0–37.4 mm³, P = 0.040, respectively] (Supplementary data online, Table S3). The time-dependent relationship for fibrofatty volume, using estimated values derived from a general linear model, is illustrated in Supplementary data online, Figure S2.

Figure 3

Lesion-level comparison of proximal plaque volumes according to early versus late ACS. Lesion-level comparison of proximal plaque volumes between culprit lesion precursors of early (blue bars, n = 62) versus late ACS (grey bars, n = 33). Precursors within the left main artery, proximal left anterior descending artery, proximal right coronary artery and proximal left circumflex artery were considered proximal. ACS, acute coronary syndrome.

Open in new tabDownload slide

Within-patient identification of future culprit lesions

As compared to within-patient non-culprit lesions, culprit lesion precursors of early ACS exhibited elevated risk for a greater maximal diameter stenosis [OR 1.050 (95% CI 1.032–1.068), P < 0.001], maximal cross-sectional plaque burden [OR 1.039 (95% CI 1.024–1.053), P < 0.001], lesion length [OR 1.033 (95% CI 1.019–1.047], P < 0.001), total plaque volume [OR 1.004 (95% CI 1.000–1.007), P = 0.032], and all compositional plaque volumes [for calcified plaque OR 1.006 (95% CI 1.000–1.011), P = 0.042; for fibrous plaque OR 1.009 (95% CI 1.002–1.017), P = 0.013; for fibrofatty plaque OR 1.010 (95% CI 1.000–1.019), P = 0.047; and for necrotic core OR 1.056 (95% CI 1.006–1.108), P = 0.029] (Table 4). Moreover, for culprit lesion precursors of late ACS an elevated risk was observed for a greater maximal diameter stenosis, lesion length, total, calcified and fibrous volumes as compared to within-patient non-culprit lesions (P < 0.05). For the significant volumetric findings, also risk measures according to tertiles were provided (Supplementary data online, Table S4).

External validation

In the external validation cohort, a total of 364 patients (mean 63 ± 9 years, 31% women) underwent baseline CCTA and experienced a MACE after a median of 531 days (IQR 31–1480 days). MACE comprised 17 (5%) MI, 10 (3%) cardiac deaths and 12 (3%) non-cardiac deaths, as well as 270 (74%) percutaneous coronary interventions and 55 (15%) coronary artery bypass grafting surgeries. Patients who had an early MACE were more often symptomatic with typical angina, dyslipidaemic, and current smokers (Supplementary data online, Table S5). As compared to patients with late MACE, patients with early MACE demonstrated significantly larger total, fibrous, fibrofatty, and necrotic core volumes [237.6 mm³ (IQR 124.4–406.8 mm³) vs. 175.3 mm³ (IQR 56.2–303.1 mm³), P < 0.001; 96.8 mm³ (IQR 45.8–164.3 mm³) vs. 69.0 mm³ (IQR 23.1–140.8 mm³), P = 0.001; 25.9 mm³ (IQR 10.0–70.5 mm³) vs. 15.8 mm³ (IQR 2.3–53.2 mm³), P = 0.003; and 1.9 mm³ (IQR 0.1–7.2 mm³) vs. 0.6 mm³ (IQR 0.0–3.9 mm³); P = 0.010, respectively] (Figure 4). Furthermore, when restricting to patients with early versus late MI or cardiac death only, trends according to compositional plaque volumes were comparable (Supplementary data online, Figure S3).

Figure 4

Patient-level comparison of plaque volumes according to early versus late MACE^a. Patient-level comparison of plaque volumes between patients with early (red bars, n = 137) versus late MACE (grey bars, n = 227). MACE, major adverse cardiac events. ^aMACE after baseline CCTA.

Open in new tabDownload slide

Discussion

The current report from the ICONIC study compared early versus late ACS risk patterns of coronary atherosclerotic plaque across the whole coronary artery tree using CCTA. We revealed that volumetric differences in composition of coronary atherosclerosis are present between patients and proximal culprit lesion precursors of ACS according to their timing antecedent to the acute event (Graphical Abstract). On a per-patient level, total and non-calcified plaque volumes were associated with early ACS, whilst on a proximal per-lesion level this was only seen for fibrofatty volumes. Importantly, patient-level findings were externally validated in the PARADIGM registry. These data suggest a potentially important role for CCTA in the future regarding the early identification of near-term plaque instability and ACS risk.

Severity and extent of coronary atherosclerosis and ACS

The Providing Regional Observations to Study Predictors of Events in the Coronary Tree (PROSPECT) study revealed, using intravascular ultrasound, that coronary lesions responsible for ACS were predominantly angiographically mild, although characterized by large cross-sectional plaque burden and a relatively small luminal area.³ In a large retrospective cohort study of 37 647 patients, Maddox et al. observed that the extent of obstructive CAD was associated with risk of MI.²² These findings were evidently supported by a higher prevalence of obstructive disease ≥50 and ≥70% amongst ACS patients versus non-ACS controls in the primary analysis of the ICONIC study (34.6% vs. 19.2%, P < 0.001; and 12.8% vs. 5.1%, P = 0.007, respectively).¹² In the present analysis, the prevalence of obstructive disease ≥50% decreased significantly from early towards late ACS, suggesting that obstructive disease was mainly associated with relatively short-term events. Preceding imaging studies have investigated the relationship of the extent of coronary atherosclerosis in stable patients with the occurrence of ACS.^2,3,23–26 Hadamitzky et al.²⁶ observed, in a 5-year follow-up study of 1584 patients, that the number of coronary segments containing any plaque predicted all-cause death and non-fatal MI in patients undergoing CCTA for suspected stable CAD. Similar findings were described in CCTA studies comparing patient-level or segment-level total plaque volumes between ACS patients and non-ACS controls.^2,24 Interestingly, the present analysis showed that the total plaque volume in patients with early ACS was larger than in those with late ACS, which may further substantiate the hypothesis that the extent of CAD is related to the temporal risk for ACS.

Composition of coronary atherosclerosis and ACS

Intravascular ultrasound studies have illustrated that echo-lucent areas within coronary lesions, likely representing necrotic core or lipid rich zones, are more frequently observed in ACS causing lesions as compared to non-ACS causing lesions.²³ Likewise, HU attenuation on CCTA has been demonstrated to depict different compositions of coronary atherosclerosis. Specific thresholds have been established to identify plaque components such as calcified plaque, fibrous plaque, fibrofatty plaque and necrotic core.⁹ Especially, the lower HU attenuations on CCTA have been acknowledged as markers for rupture-vulnerable coronary lesions with associated risk for adverse events.^2,25,27–30 Plaque rupture is the leading cause of acute thrombus formation, next to plaque erosion and to a lesser extent eruptive calcified nodules.¹⁰ The hallmark of plaque rupture is the disruption of a thin-cap fibroatheroma (<65 μm), exposing the underlying necrotic core to the lumen with subsequent luminal thrombosis.³¹ Necrotic core is a progressive stage in CAD in which inflammatory cells, such as macrophages and T lymphocytes, penetrate the lipid pool and undergo apoptosis or necrosis.³² In the current analysis using CCTA, a trend towards larger necrotic core volumes in proximal culprit lesion precursors of early versus late ACS was observed, which reached statistical significance when the volume of necrotic core was combined with fibrofatty plaque (0.5 mm³ vs. 0.1 mm³, P = 0.065; and 15.5 mm³ vs. 5.9 mm³, P = 0.040, respectively). The combination of both necrotic core and fibrofatty plaque on CCTA might reflect the lipid pool in its totality that can be observed on histopathology (gold standard).³³ Lipid pool might progress into a necrotic core during the later inflammatory stages of CAD and is therefore contributory to ACS risk. To date, it is unknown how well CCTA can distinguish lipid pool from necrotic core given the current limits of spatial resolution. To this end, future research is warranted in order to evaluate if dual-energy as compared to single-energy CCTA can further improve the characterization of plaque components.

Limitations

The present findings were part of large observational cohort studies with intrinsic limitations, including unmeasured confounding factors and selection bias. Limited information was available on duration of or changes in medication use after baseline CCTA. In 18 ACS patients long-term medications were available, showing that those with early or late ACS had similar usage of cardiac medication at 5-year follow-up. Regarding the adjudication of ACS events, potentially an important group of patients experiencing ACS in an earlier revascularized segment or leading to death without sufficient available data was excluded from the derivation cohort. Additionally, culprit lesions that caused MACE were not adjudicated in the external validation cohort (as per earlier reported study design).¹⁹ For instance, ICAs were not collected and therefore only a patient-level validation could be performed. Regarding the co-registration of culprit lesions to culprit lesion precursors, ACS patients having no or unmeasurable coronary lesions on baseline CCTA due to spatial resolution and artefacts could not be co-registered. Moreover, quantification of chronic total occlusions was not feasible with the software that was utilized. Finally, coronary atherosclerosis is a continuously dynamic process, which should always be considered when interpreting our results.

Conclusion

Volumetric differences in composition of coronary atherosclerosis exist between ACS patients experiencing an early versus late ACS after clinically indicated baseline CCTA. A large burden of non-calcified plaque across the whole coronary artery tree is strongly associated with near-term plaque instability and ACS risk, supporting the importance of plaque components related to vulnerable plaque.

Supplementary data

Supplementary data are available at European Heart Journal - Cardiovascular Imaging online.

Funding

This trial was supported by NIH Grant No. HL115150 and the Leading Foreign Research Institute Recruitment Programme of the National Research Foundation of Korea, Ministry of Science, ICT, & Future Planning (Seoul, Korea).

Data Availability

Data may be available upon reasonable request to the corresponding author.

References

Author notes