Do plaques rapidly progress prior to myocardial infarction? The interplay between plaque vulnerability and progression

Comparison of circulating dendritic cell and monocyte subsets at different stages of atherosclerosis: insights from optical coherence tomography

BACKGROUND

While specific patterns of circulating dendritic cells (DCs) and monocytes are associated with the incidence of coronary artery disease, the characterization of circulating DC and monocyte subsets in patients with different stages of atherosclerosis remains unclear.

METHODS

Forty-eight patients with unstable angina pectoris (UAP) diagnosed by angiography were enrolled. Likewise, 31 patients with ST-segment elevation myocardial infarction (STEMI) were enrolled and confirmed with the presence of thrombosis by angiography. Plaque features of 48 UAP patients were evaluated at the culprit lesions by OCT. Circulating myeloid DCs (mDCs), plasmacytoid DCs (pDCs) and monocyte subsets were analyzed using flow cytometry.

RESULTS

The proportions and absolute counts of mDC2s, which specifically express CD141 and possess the ability to activate CD8+ T lymphocytes, significantly decreased in patients with UAP and STEMI when compared with controls (0.08 × 10⁴ ± 0.05 × 104/ml and 0.08 × 10⁴ ± 0.06 × 104/ml vs. 0.11 × 10⁴ ± 0.06 × 104/ml, p = 0.027). On the other hand, patients with UAP and STEMI had significantly higher proportions and counts of Mon2 subsets. In the OCT subgroup, patients with thin-cap fibroatheroma (TCFA) had higher proportions and absolute number of Mon2 (11.96% ± 4.27% vs. 9.42% ± 4.05%, p = 0.034; 5.17 × 104/ml ± 1.92 × 104/ml vs. 3.53 × 104/ml ± 2.65 × 104/ml, p = 0.045) than those without TCFA. However, there was no remarkable difference in mDC2s between patients with and without TCFA.

CONCLUSIONS

Circulating Mon2 appears to be a promising marker for the severity of atherosclerotic plaque.

Coronary Artery Calcium Progression Is Associated With Coronary Plaque Volume Progression: Results From a Quantitative Semiautomated Coronary Artery Plaque Analysis

OBJECTIVES

The aim of this study was to determine whether coronary artery calcium (CAC) progression was associated with coronary plaque progression on coronary computed tomographic angiography.

BACKGROUND

CAC progression and coronary plaque characteristics are associated with incident coronary heart disease. However, natural history of coronary atherosclerosis has not been well described to date, and the understanding of the association between CAC progression and coronary plaque subtypes such as noncalcified plaque progression remains unclear.

METHODS

Consecutive patients who were referred to our clinic for evaluation and had serial coronary computed tomography angiography scans performed were included in the study. Coronary artery plaque (total, fibrous, fibrous-fatty, low-attenuation, densely calcified) volumes were calculated using semiautomated plaque analysis software.

RESULTS

A total of 211 patients (61.3 ± 12.7 years of age, 75.4% men) were included in the analysis. The mean interval between baseline and follow-up scans was 3.3 ± 1.7 years. CAC progression was associated with a significant linear increase in all types of coronary plaque and no plaque progression was observed in subjects without CAC progression. In multivariate analysis, annualized and normalized total plaque (β = 0.38; p < 0.001), noncalcified plaque (β = 0.35; p = 0.001), fibrous plaque (β = 0.56; p < 0.001), and calcified plaque (β = 0.63; p = 0.001) volume progression, but not fibrous-fatty (β = 0.03; p = 0.28) or low-attenuation plaque (β = 0.11; p = 0.1) progression, were independently associated with CAC progression. Plaque progression did not differ between the sexes. A significantly increased total and calcified plaque progression was observed in statin users.

CONCLUSIONS

In a clinical practice setting, progression of CAC was significantly associated with an increase in both calcified and noncalcified plaque volume, except fibrous-fatty and low-attenuation plaque. Serial CAC measurements may be helpful in determining the need for intensification of preventive treatment.

Multimodality Imaging of Angiogenesis in a Rabbit Atherosclerotic Model by GEBP11 Peptide Targeted Nanoparticles

Background and Aims: Angiogenesis is an important pathological process during progression of plaque formation, which can result in plaque hemorrhage and vulnerability. This study aims to explore non-invasive imaging of angiogenesis in atherosclerotic plaque through magnetic resonance imaging (MRI) and positron emission tomography (PET) by using GEBP11 peptide targeted magnetic iron oxide nanoparticles in a rabbit model of atherosclerosis.

Methods: The dual-modality imaging probe was constructed by coupling 2, 3-dimercaptosuccinnic acid-coated paramagnetic nanoparticles (DMSA-MNPs) and the PET ⁶⁸Ga chelator 1,4,7-triazacyclononane-N, N', N''-triacetic acid (NOTA) to GEBP11 peptide. The atherosclerosis model was induced in New Zealand white rabbits by abdominal aorta balloon de-endothelialization and atherogenic diet for 12 weeks. The plaque areas in abdominal artery were detected by ultrasound imaging and Oil Red O staining. Immunofluorescence staining and Prussian blue staining were applied respectively to investigate the affinity of GEBP11 peptide. MTT and flow cytometric analysis were performed to detect the effects of NGD-MNPs on cell proliferation, cell cycle and apoptosis in Human umbilical vein endothelial cells (HUVECs). In vivo MRI and PET imaging of atherosclerotic plaque were carried out at different time points after intravenous injection of nanoparticles.

Results: The NGD-MNPs with hydrodynamic diameter of 130.8 nm ± 7.1 nm exhibited good imaging properties, high stability, low immunogenicity and little cytotoxicity. In vivo PET/MR imaging revealed that ⁶⁸Ga-NGD-MNPs were successfully applied to visualize atherosclerotic plaque angiogenesis in the rabbit abdominal aorta. Prussian blue and CD31 immunohistochemical staining confirmed that NGD-MNPs were well co-localized within the blood vessels' plaques.

Conclusion:⁶⁸Ga-NGD-MNPs might be a promising MR and PET dual imaging probe for visualizing the vulnerable plaques.

Prognostic Significance of Nonobstructive Left Main Coronary Artery Disease in Women Versus Men: Long-Term Outcomes From the CONFIRM (Coronary CT Angiography Evaluation For Clinical Outcomes: An International Multicenter) Registry

BACKGROUND

Patients with obstructive (≥50% stenosis) left main (LM) coronary artery disease (CAD) are at high risk for adverse events; prior studies have also documented worse outcomes among women than men with severe multivessel/LM CAD. However, the prognostic significance of nonobstructive (1%-49% stenosis) LM CAD, including sex-specific differences, has not been previously examined.

METHODS AND RESULTS

In the long-term CONFIRM (Coronary CT Angiography Evaluation For Clinical Outcomes: An International Multicenter) registry, patients underwent elective coronary computed tomographic angiography for suspected CAD and were followed for 5 years. After excluding those with obstructive LM CAD, 5166 patients were categorized as having normal LM or nonobstructive LM (18% of cohort). Cumulative 5-year incidence of death, myocardial infarction, or revascularization was higher among patients with nonobstructive LM than normal LM in both women and men: women (34.3% versus 15.4%; P<0.0001); men (24.6% versus 18.2%; P<0.0001). A significant interaction existed between sex and LM status for the composite outcome (P=0.001). In multivariable Cox regression, the presence of nonobstructive LM plaque increased the risk for the composite outcome in women (adjusted hazard ratio, 1.48; P=0.005) but not in men (adjusted hazard ratio, 0.98, P=0.806). In subgroup analysis, women with nonobstructive LM CAD had a nearly 80% higher risk for events than men with nonobstructive LM CAD (adjusted hazard ratio, 1.78; P=0.017); sex-specific interactions were not observed across other patterns (eg, location or extent) of nonobstructive plaque.

CONCLUSION

Nonobstructive LM CAD was frequently detected on coronary computed tomographic angiography and strongly associated with adverse events among women. Recognizing the sex-specific prognostic significance of nonobstructive LM plaque may augment risk stratification efforts.

Very long-term serial luminal changes after sirolimus-eluting stent implantation and progression process of very late stent failure

BACKGROUND

Very long-term angiographic results after sirolimus-eluting stent (SES) implantation have not been clarified. This study investigated serial angiographic results of early (<1year), late (1-5years), and very late (>5years) follow-up after SES implantation and the progression process to very late SES failure.

METHODS

We analyzed 631 lesions undergoing serial coronary angiography at early, late, and very late follow-up after SES implantation. The results of 205 lesions undergoing very late target lesion revascularization (VL-TLR; TLR beyond 5years) for acute coronary syndrome (ACS; 83) and non-ACS (122) were evaluated in comparison with 426 non-TLR lesions as a control group.

RESULTS

Non-TLR lesions showed no attenuated decline of minimum lumen diameter from postprocedure (2.51±0.47mm), early (2.37±0.58mm), late (2.22±0.60mm), up to very late (2.01±0.63mm) follow-up. In VL-TLR lesions, compared to non-TLR lesions, late lumen loss (LLL) from postprocedure to early follow-up was similar, but delayed LLL from early to late follow-up was significantly larger (0.25±0.48mm vs. 0.15±0.46mm, p=0.01). Although the delayed LLL was significantly larger in non-ACS lesions (0.29±0.44mm) than in non-TLR lesions (p<0.01), it was similar in ACS (0.19±0.52mm) and non-TLR lesions (p=0.54). Very delayed LLL from late to very late follow-up was 1.65±0.82mm in ACS lesions and 1.10±0.76mm in non-ACS lesions.

CONCLUSIONS

Progression of in-stent luminal narrowing did not attenuate beyond 5years after SES implantation. In very late SES failure, stent-related ACS lesions showed gradual luminal narrowing and subsequent rapid progression beyond 5years, whereas non-ACS lesions had progressive luminal narrowing within 5years.

Inflammation and myointimal hyperplasia. Correlation with hemodynamic forces

OBJECTIVES

The aim of our study was to correlate flow dynamics and the release of inflammatory cytokines Interleukin 1, 2, 6, TNF (Tumour Necrosis Factor) alfa, both in vitro and in vivo.

MATERIALS AND METHODS

Endothelial cells were exposed to laminar flow (6dyne/cm²) in an in vitro circulatory system and the release of Interleukin 1, 2, 6 and TNF alfa was quantified by ELISA. Interleukin 1, 2, 6 and TNF alfa release was also assessed in vein grafts implanted in the arterial circulation of Lewis rats. Arterial vein grafts were explanted at different time intervals from 3days to 12weeks after surgery. Vein grafts implanted in the arterial circulation for 4weeks, were re implanted in the venous circulation of syngenic Lewis rats, and the release of Interleukin 1, 2, 6 and TNF alfa, was assessed in an organ culture. Six vein grafts (4 occluded, 2 patent) implanted in humans as femorodistal bypass were examined for the presence of myointimal hyperplasia and perigraft inflammatory cells.

RESULTS

In vitro, endothelial cells exposed to laminar flow released an increased amount of Interleukin 1, 2, 6 and TNF alfa in comparison to endothelial cells not exposed to flow. In experimental vein grafts implanted in the arterial circulation there was an increased release of inflammatory cytokines associated to inflammatory changes in the adventitia. Once the vein grafts were re implanted in the venous circulation, the release of these cytokines diminished, while the inflammatory changes in the adventitia regressed.

CONCLUSIONS

Increased shear stress induces release of cytokines and inflammatory changes in the adventitia. These inflammatory changes can contribute to plaque progression and to un stable plaque. These findings support the use of anti-inflammatory therapy in patients prone to develop atherosclerosis and in those who had arterial reconstructive surgery.

C1q/TNF-related protein 9 inhibits the cholesterol-induced Vascular smooth muscle cell phenotype switch and cell dysfunction by activating AMP-dependent kinase

Vascular smooth muscle cells (VSMCs) switch to macrophage‐like cells after cholesterol loading, and this change may play an important role in the progression of atherosclerosis. C1q/TNF‐related protein 9 (CTRP9) is a recently discovered adipokine that has been shown to have beneficial effects on glucose metabolism and vascular function, particularly in regard to cardiovascular disease. The question of whether CTRP9 can protect VSMCs from cholesterol damage has not been addressed. In this study, the impact of CTRP9 on cholesterol‐damaged VSMCs was observed. Our data show that in cholesterol‐treated VSMCs, CTRP9 significantly reversed the cholesterol‐induced increases in pro‐inflammatory factor secretion, monocyte adhesion, cholesterol uptake and expression of the macrophage marker CD68. Meanwhile, CTRP9 prevented the cholesterol‐induced activation of the TLR4–MyD88–p65 pathway and upregulated the expression of proteins important for cholesterol efflux. Mechanistically, as siRNA‐induced selective gene ablation of AMPKα1 abolished these effects of CTRP9, we concluded that CTRP9 achieves these protective effects in VSMCs through the AMP‐dependent kinase (AMPK) pathway.

Noninvasive Molecular Imaging of Disease Activity in Atherosclerosis

Major focus has been placed on the identification of vulnerable plaques as a means of improving the prediction of myocardial infarction. However, this strategy has recently been questioned on the basis that the majority of these individual coronary lesions do not in fact go on to cause clinical events. Attention is, therefore, shifting to alternative imaging modalities that might provide a more complete pan-coronary assessment of the atherosclerotic disease process. These include markers of disease activity with the potential to discriminate between patients with stable burnt-out disease that is no longer metabolically active and those with active atheroma, faster disease progression, and increased risk of infarction. This review will examine how novel molecular imaging approaches can provide such assessments, focusing on inflammation and microcalcification activity, the importance of these processes to coronary atherosclerosis, and the advantages and challenges posed by these techniques.

Early Detection and Treatment of the Vulnerable Coronary Plaque

Early identification and treatment of the vulnerable plaque, that is, a coronary artery lesion with a high likelihood of rupture leading to an acute coronary syndrome, have gained great interest in the cardiovascular research field. Postmortem studies have identified clear morphological characteristics associated with plaque rupture. Recent advances in invasive and noninvasive coronary imaging techniques have empowered the clinician to identify suspected vulnerable plaques in vivo and paved the way for the evaluation of therapeutic agents targeted at reducing plaque vulnerability. Local treatment of vulnerable plaques by percutaneous coronary intervention and systemic treatment with anti-inflammatory and low-density lipoprotein–lowering drugs are currently being investigated in large randomized clinical trials to assess their therapeutic potential for reducing adverse coronary events. Results from these studies may enable a more patient-tailored strategy for the treatment of coronary artery disease.

The Updated NICE Guidelines: Cardiac CT as the First-Line Test for Coronary Artery Disease

Purpose of Review

Cost-effective care pathways are integral to delivering sustainable healthcare programmes. Due to the overestimation of coronary artery disease using traditional risk tables, non-invasive testing has been utilised to improve risk stratification and initiate appropriate management to reduce the dependence on invasive investigations. In line with recent technological improvements, cardiac CT is a modality that offers a detailed anatomical assessment of coronary artery disease comparable to invasive coronary angiography.

Recent Findings

The recent publication of the National Institute for Health and Care Excellences (NICE) Clinical Guideline 95 update assesses the performance and cost utility of different non-invasive imaging strategies in patients presenting with suspected anginal chest pain. The low cost and high sensitivity of cardiac CT makes it the non-invasive test of choice in the evaluation of stable angina. This has now been ratified in national guidelines with NICE recommending cardiac CT as the first-line investigation for all patients presenting with chest pain due to suspected coronary artery disease. Additionally, randomised controlled trials have demonstrated that cardiac CT improves diagnostic certainty when incorporated into chest pain pathways.

Summary

NICE recommend cardiac CT as the first-line test for the evaluation of stable coronary artery disease in chest pain pathways.

Invasive or non-invasive imaging for detecting high-risk coronary lesions?

INTRODUCTION

Advances in our understanding about atherosclerotic evolution have enabled us to identify specific plaque characteristics that are associated with coronary plaque vulnerability and cardiovascular events. With constant improvements in signal and image processing an arsenal of invasive and non-invasive imaging modalities have been developed that are capable of identifying these features allowing in vivo assessment of plaque vulnerability. Areas covered: This review article presents the available and emerging imaging modalities introduced to assess plaque morphology and biology, describes the evidence from the first large scale studies that evaluated the efficacy of invasive and non-invasive imaging in detecting lesions that are likely to progress and cause cardiovascular events and discusses the potential implications of the in vivo assessment of coronary artery pathology in the clinical setting. Expert commentary: Invasive imaging, with its high resolution, and in particular hybrid intravascular imaging appears as the ideal approach to study the mechanisms regulating atherosclerotic disease progression; whereas non-invasive imaging is expected to enable complete assessment of coronary tree pathology, detection of high-risk lesions, more accurate risk stratification and thus to allow a personalized treatment of vulnerable patients.

High wall shear stress and high-risk plaque: an emerging concept

In recent years, there has been a significant effort to identify high-risk plaques in vivo prior to acute events. While number of imaging modalities have been developed to identify morphologic characteristics of high-risk plaques, prospective natural-history observational studies suggest that vulnerability is not solely dependent on plaque morphology and likely involves additional contributing mechanisms. High wall shear stress (WSS) has recently been proposed as one possible causative factor, promoting the development of high-risk plaques. High WSS has been shown to induce specific changes in endothelial cell behavior, exacerbating inflammation and stimulating progression of the atherosclerotic lipid core. In line with experimental and autopsy studies, several human studies have shown associations between high WSS and known morphological features of high-risk plaques. However, despite increasing evidence, there is still no longitudinal data linking high WSS to clinical events. As the interplay between atherosclerotic plaque, artery, and WSS is highly dynamic, large natural history studies of atherosclerosis that include WSS measurements are now warranted. This review will summarize the available clinical evidence on high WSS as a possible etiological mechanism underlying high-risk plaque development.

Morphological features of coronary plaques in WHHLMI rabbits (Oryctolagus cuniculus), an animal model for familial hypercholesterolemia

In order to examine their suitability for studies on coronary atherosclerosis, we evaluated the features of coronary atherosclerotic plaques in myocardial infarction-prone Watanabe heritable hyperlipidemic (WHHLMI) rabbits, a spontaneous animal model for coronary atherosclerosis and myocardial infarction. Coronary segments of the hearts of 187 WHHLMI rabbits (10–29 months old) were sectioned serially and stained histopathologically and immunohistologically. Progression of coronary lesions was prominent in rabbits that had died suddenly. The degree of coronary lesions of females was higher than that of males. Various types of atherosclerotic lesions were observed in the coronary arteries, such as plaques with a large lipid core covered by a thin fibrous cap, fatty streaks, early and advanced fibroatheromas, fibrous lesions, and advanced lesions with calcium accumulation and the vasa vasorum. In rabbits that had died suddenly, the frequencies of fibroatheromas or advanced lesions were higher than those of rabbits euthanized. Matrix metalloproteinase (MMP)-positive macrophages were detected in gaps among endothelial cells at the plaque surface, beneath the fibrous cap of thin-capped fibroatheromas, and at the bottom of the intimal plaques in which the tunica media was attenuated. Immunohistological results suggest that MMP-positive macrophages are involved in the initiation, progression, and destabilization of coronary plaques, in addition to vascular remodeling, even in WHHLMI rabbits. In conclusion, coronary lesions in WHHLMI rabbits resemble human atherosclerotic lesions, and thus, the WHHLMI rabbit is a suitable animal model for studies on human coronary plaques.

Cohort profile of BIOMArCS: the BIOMarker study to identify the Acute risk of a Coronary Syndrome-a prospective multicentre biomarker study conducted in the Netherlands

PURPOSE

Progression of stable coronary artery disease (CAD) towards acute coronary syndrome (ACS) is a dynamic and heterogeneous process with many intertwined constituents, in which a plaque destabilising sequence could lead to ACS within short time frames. Current CAD risk assessment models, however, are not designed to identify increased vulnerability for the occurrence of coronary events within a precise, short time frame at the individual patient level. The BIOMarker study to identify the Acute risk of a Coronary Syndrome (BIOMArCS) was designed to evaluate whether repeated measurements of multiple biomarkers can predict such 'vulnerable periods'.

PARTICIPANTS

BIOMArCS is a multicentre, prospective, observational study of 844 patients presenting with ACS, either with or without ST-elevation and at least one additional cardiovascular risk factor.

METHODS AND ANALYSIS

We hypothesised that patterns of circulating biomarkers that reflect the various pathophysiological components of CAD, such as distorted lipid metabolism, vascular inflammation, endothelial dysfunction, increased thrombogenicity and ischaemia, diverge in the days to weeks before a coronary event. Divergent biomarker patterns, identified by serial biomarker measurements during 1-year follow-up might then indicate 'vulnerable periods' during which patients with CAD are at high short-term risk of developing an ACS. Venepuncture was performed every fortnight during the first half-year and monthly thereafter. As prespecified, patient enrolment was terminated after the primary end point of cardiovascular death or hospital admission for non-fatal ACS had occurred in 50 patients. A case-cohort design will explore differences in temporal patterns of circulating biomarkers prior to the repeat ACS.

FUTURE PLANS AND DISSEMINATION

Follow-up and event adjudication have been completed. Prespecified biomarker analyses are currently being performed and dissemination through peer-reviewed publications and conference presentations is expected from the third quarter of 2016. Should identification of a 'vulnerable period' prove to be feasible, then future research could focus on event reduction through pharmacological or mechanical intervention during such periods of high risk for ACS.

TRIAL REGISTRATION NUMBER

NTR1698 and NTR1106.

Presence of Severe Stenosis in Most Culprit Lesions of Patients with ST-segment Elevation Myocardial Infarction

Background:

Previous studies revealed that culprit vessels of ST-segment elevation myocardial infarction (STEMI) were often related to mild or moderate stenosis. However, recent studies suggested that severe stenosis was primarily found in culprit lesions. The objective of this study was to analyze the stenosis severity of culprit lesions in STEMI patients and to clarify the paradoxical results.

Methods:

A total of 489 consecutive STEMI patients who underwent primary percutaneous coronary intervention were retrospectively studied from January 2012 to December 2014. The patients were divided into three groups based on stenosis severity using quantitative coronary analysis: Group A, 314 cases, stenosis ≥70%; Group B, 127 cases, stenosis 50–70%; and Group C, 48 cases, stenosis ≤50%. The clinical, demographic, and angiographic data of all groups were analyzed.

Results:

Patients in Group A exhibited a significantly higher prevalence of history of angina pectoris (95.9% vs. 62.5%, P < 0.001), multivessel disease (73.2% vs. 54.2%, P = 0.007), and lower cardiac ejection fraction (53.3 ± 8.6 vs. 56.8 ± 8.4, P = 0.009) than those in Group C. Multivariable analysis revealed that history of angina pectoris (odds ratio [OR]: 13.89, 95% confidence interval [CI]: 6.21–31.11) and multivessel disease (OR: 2.32, 95% CI: 1.25–4.31) were correlated with severe stenosis of the culprit lesion in Group A.

Conclusions:

Most culprit lesions in STEMI patients were severe stenosis. These patients exhibited a higher prevalence of angina history and multivessel diseases.

Prognostic Determinants of Coronary Atherosclerosis in Stable Ischemic Heart Disease

Risk stratification in patients with stable ischemic heart disease is essential to guide treatment decisions. In this regard, whether coronary anatomy, physiology, or plaque morphology is the best determinant of prognosis (and driver an effective therapeutic risk reduction) remains one of the greatest ongoing debates in cardiology. In the present report, we review the evidence for each of these characteristics and explore potential algorithms that may enable a practical diagnostic and therapeutic strategy for the management of patients with stable ischemic heart disease.

Imaging Atherosclerosis

Advances in atherosclerosis imaging technology and research have provided a range of diagnostic tools to characterize high-risk plaque in vivo; however, these important vascular imaging methods additionally promise great scientific and translational applications beyond this quest. When combined with conventional anatomic- and hemodynamic-based assessments of disease severity, cross-sectional multimodal imaging incorporating molecular probes and other novel noninvasive techniques can add detailed interrogation of plaque composition, activity, and overall disease burden. In the catheterization laboratory, intravascular imaging provides unparalleled access to the world beneath the plaque surface, allowing tissue characterization and measurement of cap thickness with micrometer spatial resolution. Atherosclerosis imaging captures key data that reveal snapshots into underlying biology, which can test our understanding of fundamental research questions and shape our approach toward patient management. Imaging can also be used to quantify response to therapeutic interventions and ultimately help predict cardiovascular risk. Although there are undeniable barriers to clinical translation, many of these hold-ups might soon be surpassed by rapidly evolving innovations to improve image acquisition, coregistration, motion correction, and reduce radiation exposure. This article provides a comprehensive review of current and experimental atherosclerosis imaging methods and their uses in research and potential for translation to the clinic.

PCA-based polling strategy in machine learning framework for coronary artery disease risk assessment in intravascular ultrasound: A link between carotid and coronary grayscale plaque morphology

BACKGROUND AND OBJECTIVE

Percutaneous coronary interventional procedures need advance planning prior to stenting or an endarterectomy. Cardiologists use intravascular ultrasound (IVUS) for screening, risk assessment and stratification of coronary artery disease (CAD). We hypothesize that plaque components are vulnerable to rupture due to plaque progression. Currently, there are no standard grayscale IVUS tools for risk assessment of plaque rupture. This paper presents a novel strategy for risk stratification based on plaque morphology embedded with principal component analysis (PCA) for plaque feature dimensionality reduction and dominant feature selection technique. The risk assessment utilizes 56 grayscale coronary features in a machine learning framework while linking information from carotid and coronary plaque burdens due to their common genetic makeup.

METHOD

This system consists of a machine learning paradigm which uses a support vector machine (SVM) combined with PCA for optimal and dominant coronary artery morphological feature extraction. Carotid artery proven intima-media thickness (cIMT) biomarker is adapted as a gold standard during the training phase of the machine learning system. For the performance evaluation, K-fold cross validation protocol is adapted with 20 trials per fold. For choosing the dominant features out of the 56 grayscale features, a polling strategy of PCA is adapted where the original value of the features is unaltered. Different protocols are designed for establishing the stability and reliability criteria of the coronary risk assessment system (cRAS).

RESULTS

Using the PCA-based machine learning paradigm and cross-validation protocol, a classification accuracy of 98.43% (AUC 0.98) with K=10 folds using an SVM radial basis function (RBF) kernel was achieved. A reliability index of 97.32% and machine learning stability criteria of 5% were met for the cRAS.

CONCLUSIONS

This is the first Computer aided design (CADx) system of its kind that is able to demonstrate the ability of coronary risk assessment and stratification while demonstrating a successful design of the machine learning system based on our assumptions.