Clinical expert consensus document on quantitative coronary angiography from the Japanese Association of Cardiovascular Intervention and Therapeutics.Cardiovasc Interv Ther. 2020;35:105-116. [PMID: 32125622 DOI: 10.1007/s12928-020-00653-7]

Future Horizons: The Potential Role of Artificial Intelligence in Cardiology

Cardiovascular diseases (CVDs) are the leading cause of premature death and disability globally, leading to significant increases in healthcare costs and economic strains. Artificial intelligence (AI) is emerging as a crucial technology in this context, promising to have a significant impact on the management of CVDs. A wide range of methods can be used to develop effective models for medical applications, encompassing everything from predicting and diagnosing diseases to determining the most suitable treatment for individual patients. This literature review synthesizes findings from multiple studies that apply AI technologies such as machine learning algorithms and neural networks to electrocardiograms, echocardiography, coronary angiography, computed tomography, and cardiac magnetic resonance imaging. A narrative review of 127 articles identified 31 papers that were directly relevant to the research, encompassing a broad spectrum of AI applications in cardiology. These applications included AI models for ECG, echocardiography, coronary angiography, computed tomography, and cardiac MRI aimed at diagnosing various cardiovascular diseases such as coronary artery disease, hypertrophic cardiomyopathy, arrhythmias, pulmonary embolism, and valvulopathies. The papers also explored new methods for cardiovascular risk assessment, automated measurements, and optimizing treatment strategies, demonstrating the benefits of AI technologies in cardiology. In conclusion, the integration of artificial intelligence (AI) in cardiology promises substantial advancements in diagnosing and treating cardiovascular diseases.

Tandem lesions associate with angiographic progression of coronary artery stenoses

Background

Although the clinical factors associated with progression of coronary artery disease have been well studied, the angiographic predictors are less defined.

Objectives

Our objective was to study the clinical and angiographic factors that associate with progression of coronary artery stenoses.

Methods

We conducted a retrospective analysis of consecutive patients undergoing multiple, clinically indicated invasive coronary angiograms with an interval greater than 6 months, between January 2013 and December 2016. Lesion segments were analysed using Quantitative Coronary Angiography (QCA) if a stenosis ≥ 20 % was identified on either angiogram. Stenosis progression was defined as an increase ≥ 10 % in stenosis severity, with progressor groups analysed on both patient and lesion levels. Mixed-effects regression analyses were performed to evaluate factors associated with progression of individual stenoses.

Results

199 patients were included with 881 lesions analysed. 108 (54.3 %) patients and 186 (21.1 %) stenoses were classified as progressors. The median age was 65 years (IQR 56-73) and the median interval between angiograms was 2.1 years (IQR 1.2-3.0). On a patient level, age, number of lesions and presence of multivessel disease at baseline were each associated with progressor status. On a lesion level, presence of a stenosis downstream (OR 3.07, 95 % CI 2.04-4.63, p < 0.001) and circumflex artery stenosis location (OR 1.81, 95 % CI 1.21-2.7, p = 0.004) were associated with progressor status. Other lesion characteristics did not significantly impact progressor status or change in stenosis severity.

Conclusion

Coronary lesions which have a downstream stenosis may be at increased risk of stenosis progression. Further research into the mechanistic basis of this finding is required, along with its implications for plaque vulnerability and clinical outcomes.

Predictors of late lumen enlargement after drug-coated balloon angioplasty for de novo coronary lesions

BACKGROUND

Late lumen enlargement (LLE) - a positive remodelling phenomenon - after drug-coated balloon (DCB) angioplasty for stable coronary disease contributes to a lower restenosis rate. However, lesion characteristics promoting LLE remain unclear.

AIMS

This study aimed to investigate predictive lesion characteristics for LLE using serial optical frequency domain imaging (OFDI) following DCB angioplasty for de novo coronary artery lesions.

METHODS

This retrospective, single-centre observational study included patients with angina pectoris who underwent paclitaxel-coated balloon angioplasty without stenting under OFDI guidance as well as follow-up OFDI. OFDI endpoints were lumen volume, plaque phenotype, and procedure-associated dissection. LLE was defined as a ≥10% increase in the lumen volume of the treated lesion at follow-up.

RESULTS

Between August 2016 and December 2019, among patients with successful DCB angioplasty, 108 lesions (83 patients) had available follow-up imaging after a median of 6.1 months. LLE was detected in 44 (40.7%) lesions. Fibrous/fibrocalcific and layered plaques had significantly larger lumen volumes at follow-up than immediately after the index procedure, whereas lipid plaques exhibited no significant difference. Medial dissection with an arc >90° revealed an increased lumen volume. Multivariate analysis showed that layered plaques (odds ratio [OR] 8.73, 95% confidence interval [CI]: 1.92-39.7; p=0.005) and medial dissection with an arc >90° (OR 4.65, 95% CI: 1.63-13.3; p=0.004) were independent LLE predictors.

CONCLUSIONS

Layered plaques and extensive medial dissection after DCB angioplasty were associated with higher LLE occurrence in de novo coronary lesions. These findings may be clinically applicable to DCB therapeutic strategies based on plaque features.

Fully automated quantitative coronary angiography versus optical coherence tomography guidance for coronary stent implantation (FLASH): Study protocol for a randomized controlled noninferiority trial

BACKGROUND

Artificial intelligence-based quantitative coronary angiography (AI-QCA) has been developed to provide a more objective and reproducible data about the severity of coronary artery stenosis and the dimensions of the vessel for intervention in real-time, overcoming the limitations of significant inter- and intraobserver variability, and time-consuming nature of on-site QCA, without requiring extra time and effort. Compared with the subjective nature of visually estimated conventional CAG guidance, AI-QCA guidance provides a more practical and standardized angiography-based approach. Although the advantage of intravascular imaging-guided PCI is increasingly recognized, their broader adoption is limited by clinical and economic barriers in many catheterization laboratories.

METHODS

The FLASH (fully automated quantitative coronary angiography versus optical coherence tomography guidance for coronary stent implantation) trial is a randomized, investigator-initiated, multicenter, open-label, noninferiority trial comparing the AI-QCA-assisted PCI strategy with optical coherence tomography-guided PCI strategy in patients with significant coronary artery disease. All operators will utilize a novel, standardized AI-QCA software and PCI protocol in the AI-QCA-assisted group. A total of 400 patients will be randomized to either group at a 1:1 ratio. The primary endpoint is the minimal stent area (mm2), determined by the final OCT run after completion of PCI. Clinical follow-up and cost-effectiveness evaluations are planned at 1 month and 6 months for all patients enrolled in the study.

RESULTS

Enrollment of a total of 400 patients from the 13 participating centers in South Korea will be completed in February 2024. Follow-up of the last enrolled patients will be completed in August 2024, and primary results will be available by late 2024.

CONCLUSION

The FLASH is the first clinical trial to evaluate the feasibility of AI-QCA-assisted PCI, and will provide the clinical evidence on AI-QCA assistance in the field of coronary intervention.

CLINICAL TRIAL REGISTRATION

URL: https://www.

CLINICALTRIALS

gov. Unique identifier: NCT05388357.

Quantitative Angiography: The Dawn of a New Era in Cardiovascular Medicine

This comprehensive review explores the transformative role of quantitative angiography in the landscape of cardiovascular medicine. Tracing the historical evolution of cardiovascular diagnostics, we emphasize the significance of angiography in diagnosis and treatment. The primary focus on quantitative angiography reveals its capacity to move beyond qualitative assessments, providing clinicians with precise measurements and objective parameters. This paradigm shift enhances diagnostic accuracy, promising far-reaching implications for the future of cardiovascular medicine. The ability to tailor interventions based on meticulous measurements optimizes therapeutic strategies and positions the field on the brink of a new era where personalized approaches become the norm. However, challenges such as image quality, radiation exposure, and interpretation variability persist, necessitating a collective call to action for continued research and development. As we confront these issues, collaborative efforts across disciplines are essential to refine existing technologies and usher in innovative solutions. This review concludes with a resounding call for ongoing research initiatives, large-scale clinical studies, and collective commitment to propel quantitative angiography into a universally accepted standard, ensuring its full realization in enhancing patient care and outcomes in cardiovascular medicine.

Impact of baseline yellow plaque assessed by coronary angioscopy on vascular response after stent implantation

BACKGROUND

The relationship between baseline yellow plaque (YP) and vascular response after stent implantation has not been fully investigated.

METHODS

This was a sub-analysis of the Collaboration-1 study (multicenter, retrospective, observational study). A total of 88 lesions from 80 patients with chronic coronary syndrome who underwent percutaneous coronary intervention were analyzed. Optical coherence tomography (OCT) and coronary angioscopy (CAS) were serially performed immediately and 11 months after stent implantation. YP was defined as the stented segment with yellow or intensive yellow color assessed by CAS. Neoatherosclerosis was defined as a lipid or calcified neointima assessed by OCT. OCT and CAS findings at 11 months were compared between lesions with baseline YP (YP group) and lesions without baseline YP (Non-YP group).

RESULTS

Baseline YP was detected in 37 lesions (42 %). OCT findings at 11 months showed that the incidence of neoatherosclerosis was significantly higher in the YP group (11 % versus 0 %, p = 0.028) and mean neointimal thickness tended to be lower (104 ± 43 μm versus 120 ± 48 μm, p = 0.098). CAS findings at 11 months demonstrated that the dominant and minimum neointimal coverage grades were significantly lower (p = 0.049 and P = 0.026) and maximum yellow color grade was significantly higher (p < 0.001) in the YP group.

CONCLUSIONS

Baseline YP affected the incidence of neoatherosclerosis as well as poor neointimal coverage at 11 months after stent implantation.

Influence of intracoronary hemodynamic forces on atherosclerotic plaque phenotypes

BACKGROUND AND AIMS

Coronary hemodynamics impact coronary plaque progression and destabilization. The aim of the present study was to establish the association between focal vs. diffuse intracoronary pressure gradients and wall shear stress (WSS) patterns with atherosclerotic plaque composition.

METHODS

Prospective, international, single-arm study of patients with chronic coronary syndromes and hemodynamic significant lesions (fractional flow reserve [FFR] ≤ 0.80). Motorized FFR pullback pressure gradient (PPG), optical coherence tomography (OCT), and time-average WSS (TAWSS) and topological shear variation index (TSVI) derived from three-dimensional angiography were obtained.

RESULTS

One hundred five vessels (median FFR 0.70 [Interquartile range (IQR) 0.56-0.77]) had combined PPG and WSS analyses. TSVI was correlated with PPG (r = 0.47, [95% Confidence Interval (95% CI) 0.30-0.65], p < 0.001). Vessels with a focal CAD (PPG above the median value of 0.67) had significantly higher TAWSS (14.8 [IQR 8.6-24.3] vs. 7.03 [4.8-11.7] Pa, p < 0.001) and TSVI (163.9 [117.6-249.2] vs. 76.8 [23.1-140.9] m-1, p < 0.001). In the 51 vessels with baseline OCT, TSVI was associated with plaque rupture (OR 1.01 [1.00-1.02], p = 0.024), PPG with the extension of lipids (OR 7.78 [6.19-9.77], p = 0.003), with the presence of thin-cap fibroatheroma (OR 2.85 [1.11-7.83], p = 0.024) and plaque rupture (OR 4.94 [1.82 to 13.47], p = 0.002).

CONCLUSIONS

Focal and diffuse coronary artery disease, defined using coronary physiology, are associated with differential WSS profiles. Pullback pressure gradients and WSS profiles are associated with atherosclerotic plaque phenotypes. Focal disease (as identified by high PPG) and high TSVI are associated with high-risk plaque features.

CLINICAL TRIAL REGISTRATION

https://clinicaltrials,gov/ct2/show/NCT03782688.

Association Between Automated 3D Measurement of Coronary Luminal Narrowing and Risk of Future Myocardial Infarction

This study focuses on identifying anatomical markers with predictive capacity for long-term myocardial infarction (MI) in focal coronary artery disease (CAD). Eighty future culprit lesions (FCL) and 108 non-culprit lesions (NCL) from 80 patients underwent 3D quantitative coronary angiography. The minimum lumen area (MLA), minimum lumen ratio (MLR), and vessel fractional flow reserve (vFFR) were evaluated. MLR was defined as the ratio between MLA and the cross-sectional area at the proximal lesion edge, with lower values indicating more abrupt luminal narrowing. Significant differences were observed between FCL and NCL in MLR (0.41 vs. 0.53, p < 0.001). MLR correlated inversely with translesional vFFR (r =  - 0.26, p = 0.0004) and was the strongest predictor of MI at 5 years (AUC = 0.75). Lesions with MLR < 0.40 had a fourfold increased MI incidence at 5 years. MLR is a robust predictor of future adverse coronary events.

A novel coronary angiographic index for predicting correlation between fractional flow reserve and resting full-cycle ratio

OBJECTIVES

The discordant results between fractional flow reserve (FFR) and resting full-cycle ratio (RFR) and the influence of angiographic characteristics on their correlation have not been sufficiently investigated. We aimed to identify angiographic characteristics that can predict FFR and RFR correlations using a novel angiographic scoring system.

METHODS

This retrospective analysis included 220 patients with 252 intermediate coronary lesions assessed using FFR and RFR. Each branch distal to the target lesion was scored based on the vessel diameter (0 points: < 1.5 mm, 1 point: 1.5-2.0 mm, and 2 points: > 2.0 mm) measured using quantitative coronary angiography. The angiographic score was calculated by adding these scores.

RESULTS

In a propensity score-matched cohort including 84 lesions (42 lesions in each low-and high-angiographic score group), the correlation between FFR and RFR in the high-angiographic score group (>4) was weaker than that in the low-score group (≤4) (Spearman's correlation: r = 0.44 vs. r = 0.80, P  < 0.01). Considering a threshold of functional myocardial ischemia as FFR ≤ 0.80 and RFR ≤ 0.89, the low-angiographic score group showed a significantly lower discordance rate of abnormal FFR/normal RFR than the high-angiographic score group (7.1% vs. 23.8%, P  = 0.03), whereas the discordance rates of normal FFR/abnormal RFR were similar in both groups (7.1% vs. 9.5%, P  = 0.69).

CONCLUSION

This retrospective analysis highlights the influence of angiographic characteristics on the correlation between FFR and RFR. Our simple angiographic assessment method may be useful for interpreting physiological evaluations in daily clinical practice.

Very short-term tissue coverage of the CD34 antibody-covered sirolimus-eluting stent: an optical coherence tomography study

It is well known that patients with acute myocardial infarction (AMI) have delayed vessel healing despite accelerated endothelial progenitor cells (EPC) mobilization. The COMBO stent is a unique biodegradable-polymer sirolimus-eluting stent with an anti-CD34 antibody coating which captures EPC and potentially promotes vessel healing. However, there are limited data about strut tissue coverage at the very short-term period after COMBO stent implantation. This was a prospective study to investigate strut tissue coverage within 1 month after COMBO stent implantation using optical coherence tomography (OCT). Struts fully covered with tissue were defined as covered, and struts with distance from lumen surface longer than strut plus polymer thickness were defined as malapposed. Mean tissue thickness was measured only in apposed struts. A total of 8173 struts of 33 lesions in 32 patients were analyzed at an average of 19.8 ± 4.6 days after COMBO stent implantation. In lesion-level analysis, the rate of covered struts was 89.6 ± 7.2%, the rate of malapposed struts was 0.9 ± 2.0% and mean tissue thickness was 46.8 ± 14.3 µm. In comparison between AMI (n = 12) and non-AMI (n = 21) patients, there were no significant differences in the rate of covered struts (88.4 ± 8.4% vs. 90.2 ± 6.6%, p = 0.48) and mean tissue thickness (46.8 ± 13.7 µm vs. 46.9 ± 15.0 µm, p = 0.98). Multivariable analysis demonstrated that time from implantation to OCT imaging was significantly associated with mean tissue thickness. The COMBO stent had substantial tissue coverage at the very short-term period after implantation even in AMI patients, and follow-up time had an impact on vessel healing.

Segmentation of X-ray coronary angiography with an artificial intelligence deep learning model: Impact in operator visual assessment of coronary stenosis severity

BACKGROUND

Visual assessment of the percentage diameter stenosis (%DSVE ) of lesions is essential in coronary angiography (CAG) interpretation. We have previously developed an artificial intelligence (AI) model capable of accurate CAG segmentation. We aim to compare operators' %DSVE in angiography versus AI-segmented images.

METHODS

Quantitative coronary analysis (QCA) %DS (%DSQCA ) was previously performed in our published validation dataset. Operators were asked to estimate %DSVE of lesions in angiography versus AI-segmented images in separate sessions and differences were assessed using angiography %DSQCA as reference.

RESULTS

A total of 123 lesions were included. %DSVE was significantly higher in both the angiography (77% ± 20% vs. 56% ± 13%, p < 0.001) and segmentation groups (59% ± 20% vs. 56% ± 13%, p < 0.001), with a much smaller absolute %DS difference in the latter. For lesions with %DSQCA of 50%-70% (60% ± 5%), an even higher discrepancy was found (angiography: 83% ± 13% vs. 60% ± 5%, p < 0.001; segmentation: 63% ± 15% vs. 60% ± 5%, p < 0.001). Similar, less pronounced, findings were observed for %DSQCA  < 50% lesions, but not %DSQCA  > 70% lesions. Agreement between %DSQCA /%DSVE across %DSQCA strata (<50%, 50%-70%, >70%) was approximately twice in the segmentation group (60.4% vs. 30.1%; p < 0.001). %DSVE inter-operator differences were smaller with segmentation.

CONCLUSION

%DSVE was much less discrepant with segmentation versus angiography. Overestimation of %DSQCA  < 70% lesions with angiography was especially common. Segmentation may reduce %DSVE overestimation and thus unwarranted revascularization.

Accelerated stenotic flow in the left anterior descending coronary artery explains the causes of impaired coronary flow reserve: an integrated transthoracic enhanced Doppler study

Background

Accelerated stenotic flow (AsF) in the entire left anterior descending coronary artery (LAD), assessed by transthoracic enhanced color Doppler (E-Doppler TTE), can reveal coronary stenosis (CS) and its severity, enabling a distinction between the microcirculatory and epicardial causes of coronary flow reserve (CFR) impairment.

Methods

Eighty-four consecutive patients with a CFR <2.0 (1.5 ± 0.4), as assessed by E-Doppler TTE, scheduled for coronary angiography (CA) and eventually intracoronary ultrasounds (IVUS), were studied. CFR was calculated by the ratio of peak diastolic flow velocities: during i.v. adenosine (140 mcg/Kg/m) over resting; AsF was calculated as the percentage increase of localized maximal velocity in relation to a reference velocity.

Results

CA showed ≥50% lumen diameter narrowing of the LAD (critical CS) in 68% of patients (57/84) vs. non-critical CS in 32% (27/84). Based on the established CA/IVUS criteria, the non-critical CS subgroup was further subdivided into 2 groups: subcritical/diffuse [16/27 pts (57%)] and no atherosclerosis [11/27 pts (43%)]. CFR was similar in the three groups: 1.4 ± 0.3 in critical CS, 1.5 ± 0.4 in subcritical/diffuse CS, and 1.6 ± 0.4 in no atherosclerosis (p = ns). Overall, at least one segment of accelerated stenotic flow in the LAD was found in 73 patients (87%), while in 11 (13%) it was not. The AsF was very predictive of coronary segmental narrowing in both angio subgroups of atherosclerosis but as expected with the usage of different cutoffs. On the basis of the ROC curve, the optimal cutoff was 109% and 16% AsF % increment to successfully distinguish critical from non-critical CS (area under the curve [AUC] = 0.99, p < 0.001) and diffuse/subcritical from no CS (AUC = 0.91%, p < 0.001). Sensitivity and specificity were 96% and 100% and 82% and 100%, respectively.

Conclusion

E-Doppler TTE is highly feasible and reliable in detecting the CS of any grade of severity, distinguishing epicardial athero from microvascular causes of a severe CFR reduction.

Assessment of the Severity of Left Anterior Descending Coronary Artery Stenoses by Enhanced Transthoracic Doppler Echocardiography: Validation of a Method Based on the Continuity Equation

BACKGROUND

To verify whether the severity of coronary stenosis could be non-invasively assessed by enhanced transthoracic coronary echo Doppler in convergent color Doppler mode (E-Doppler TTE) over a wide range of values (from mild to severe).

METHODS

Color-guided pulsed wave Doppler sampling in the left anterior descending coronary artery (LAD) was performed in 103 diseased LAD segments (corresponding to 94 patients examined) as assessed by quantitative coronary angiography (QCA) or intracoronary ultrasound (IVUS). The E-Doppler TTE examinations consisted of measuring the velocity (vel) at the stenosis site and a reference adjacent segment. Then the continuity equation (C-Eq) was applied to calculate the percent cross-sectional area reduction (%CSA) at the stenosis site. The applied formula was: %CSA = 100 × (1 - [TVIref × 0.5]/TVIs). TVI = the time velocity integral at the stenosis [s] and the reference site [ref], respectively); 0.5 = the correcting factor for a parabolic profile was used only when the % accelerated stenotic flow was >122% (AsF = diastolic peak vel at first site - diastolic peak vel at second site/diastolic peak vel at second site × 100).

RESULTS

E-Doppler TTE feasibility was 100%. Doppler and QCA/IVUS-derived %CSA stenosis showed very good agreement over a large range of values (from mild to severe), with no significant bias; the maximum difference between QCA/IVUS and transthoracic Doppler %CSA was mostly around 20% with a few patients exceeding this limit (limits of agreement = -27.53 to 23.5%). The scattering was slightly larger for the non-significant stenoses. The correlation was strong (r = 0.89, p < 0.001).

CONCLUSION

E-Doppler TTE is a feasible and reliable method for assessing the severity of LAD stenosis by applying the C-Eq.

The Application of Deep Learning for the Segmentation and Classification of Coronary Arteries

In recent years, the prevalence of coronary artery disease (CAD) has become one of the leading causes of death around the world. Accurate stenosis detection of coronary arteries is crucial for timely treatment. Cardiologists use visual estimations when reading coronary angiography images to diagnose stenosis. As a result, they face various challenges which include high workloads, long processing times and human error. Computer-aided segmentation and classification of coronary arteries, as to whether stenosis is present or not, significantly reduces the workload of cardiologists and human errors caused by manual processes. Moreover, deep learning techniques have been shown to aid medical experts in diagnosing diseases using biomedical imaging. Thus, this study proposes the use of automatic segmentation of coronary arteries using U-Net, ResUNet-a, UNet++, models and classification using DenseNet201, EfficientNet-B0, Mobilenet-v2, ResNet101 and Xception models. In the case of segmentation, the comparative analysis of the three models has shown that U-Net achieved the highest score with a 0.8467 Dice score and 0.7454 Jaccard Index in comparison with UNet++ and ResUnet-a. Evaluation of the classification model's performances has shown that DenseNet201 performed better than other pretrained models with 0.9000 accuracy, 0.9833 specificity, 0.9556 PPV, 0.7746 Cohen's Kappa and 0.9694 Area Under the Curve (AUC).

Intravascular Ultrasound Guided Intervention in Calcified Coronary Lesions Showed Good Clinical Outcomes during One Year Follow-Up

BACKGROUND

Calcified coronary lesions can cause stent under-expansion, malapposition, and polymer degradation, hence increasing the risk of adverse clinical outcomes. Percutaneous coronary intervention (PCI) guided by intravascular ultrasound (IVUS) has been used regularly to improve outcomes. Our primary aim was to evaluate the clinical efficacy of IVUS-guided PCI in calcified coronary lesions.

METHODS

From August 2018 to December 2021, we prospectively included 300 patients in the CAPIRO study (CAlcified plaque in patients receiving Resolute Onyx^®) at three educational hospitals in Jeonbuk Province. We studied 243 patients (265 lesions) who were followed up for over a year. Based on coronary calcification by IVUS analysis, the patient population was categorized into two groups (Group I: non/mild calcification; Group II: moderate/severe calcification (maximum calcium arc >180° and calcium length > 5 mm)). One-to-one Propensity Score Matching was used to match the baseline characteristics. The stent expansion rate was analyzed by recent criteria. The primary clinical outcome was Major Adverse Cardiac Events (MACE), which included Cardiac death, Myocardial Infarction (MI), and Target Lesion Revascularization (TLR).

RESULTS

After follow-up time, the MACE rate in Group I was 1.99%, comparable to Group II's 1.09% (p = 0.594). The components of MACE did not significantly differ between the two groups. Based on absolute MSA or MSA/MVA at MSA site criteria, the stent expansion rate in Group II was lower than that of Group I. Nevertheless, based on recent relative criteria, the stent expansion rate in both groups was comparable.

CONCLUSIONS

After more than a year of follow-up, IVUS-guided PCI in moderate/severe calcification lesions was associated with good clinical outcomes, which was comparable with non/mild calcification lesions. Future studies with a larger sample size and a more extended follow-up period are required to clarify our findings.

Coronary X-ray angiography segmentation using Artificial Intelligence: a multicentric validation study of a deep learning model

INTRODUCTION

We previously developed an artificial intelligence (AI) model for automatic coronary angiography (CAG) segmentation, using deep learning. To validate this approach, the model was applied to a new dataset and results are reported.

METHODS

Retrospective selection of patients undergoing CAG and percutaneous coronary intervention or invasive physiology assessment over a one month period from four centers. A single frame was selected from images containing a lesion with a 50-99% stenosis (visual estimation). Automatic Quantitative Coronary Analysis (QCA) was performed with a validated software. Images were then segmented by the AI model. Lesion diameters, area overlap [based on true positive (TP) and true negative (TN) pixels] and a global segmentation score (GSS - 0 -100 points) - previously developed and published - were measured.

RESULTS

123 regions of interest from 117 images across 90 patients were included. There were no significant differences between lesion diameter, percentage diameter stenosis and distal border diameter between the original/segmented images. There was a statistically significant albeit minor difference [0,19 mm (0,09-0,28)] regarding proximal border diameter. Overlap accuracy ((TP + TN)/(TP + TN + FP + FN)), sensitivity (TP / (TP + FN)) and Dice Score (2TP / (2TP + FN + FP)) between original/segmented images was 99,9%, 95,1% and 94,8%, respectively. The GSS was 92 (87-96), similar to the previously obtained value in the training dataset.

CONCLUSION

the AI model was capable of accurate CAG segmentation across multiple performance metrics, when applied to a multicentric validation dataset. This paves the way for future research on its clinical uses.

Fractional flow reserve and non-hyperemic indices: Essential tools for percutaneous coronary interventions

Hemodynamical evaluation of a coronary artery lesion is an important diagnostic step to assess its functional impact. Fractional flow reserve (FFR) received a class IA recommendation from the European Society of Cardiology for the assessment of angiographically moderate stenosis. FFR evaluation of coronary artery disease offers improvement of the therapeutic strategy, deferring unnecessary procedures for lesions with a FFR > 0.8, improving patients' management and clinical outcome. Post intervention, an optimal FFR > 0.9 post stenting should be reached and > 0.8 post drug eluting balloons. Non-hyperemic pressure ratio measurements have been validated in previous studies with a common threshold of 0.89. They might overestimate the hemodynamic significance of some lesions but remain useful whenever hyperemic agents are contraindicated. FFR remains the gold standard reference for invasive assessment of ischemia. We illustrate this review with two cases introducing the possibility to estimate also non-invasively FFR from reconstructed 3-D angiograms by quantitative flow ratio. We conclude introducing a hybrid approach to intermediate lesions (DFR 0.85-0.95) potentially maximizing clinical decision from all measurements.

Progressive Perception Learning for Main Coronary Segmentation in X-Ray Angiography

Main coronary segmentation from the X-ray angiography images is important for the computer-aided diagnosis and treatment of coronary disease. However, it confronts the challenge at three different image granularities (the semantic, surrounding, and local levels). The challenge includes the semantic confusion between the main and collateral vessels, low contrast between the foreground vessel and background surroundings, and local ambiguity near the vessel boundaries. The traditional hand-crafted feature-based methods may be insufficient because they may lack the semantic relationship information and may not distinguish the main and collateral vessels. The existing deep learning-based methods seem to have issues due to the deficiency in the long-distance semantic relationship capture, the foreground and background interference adaptability, and the boundary detail information preservation. To solve the main coronary segmentation challenge, we propose the progressive perception learning (PPL) framework to inspect these three different image granularities. Specifically, the PPL contains the context, interference, and boundary perception modules. The context perception is designed to focus on the main coronary vessel based on the semantic dependence capture among different coronary segments. The interference perception is designed to purify the feature maps based on the foreground vessel enhancement and background artifact suppression. The boundary perception is designed to highlight the boundary details based on boundary feature extraction through the intersection between the foreground and background predictions. Extensive experiments on 1085 subjects show that the PPL is effective (e.g., the overall Dice is greater than 95%), and superior to thirteen state-of-the-art coronary segmentation methods.

Relation of GRACE Risk Score to Coronary Lipid Core Plaques in Patients with Acute Coronary Syndrome

The GRACE risk score is established to predict thrombotic events in patients with acute coronary syndrome (ACS). Although thrombotic events including myocardial infarction after ACS are mainly attributable to vulnerable plaque formation, whether the GRACE score correlates with coronary lipid-rich plaque is unclear. A total of 54 patients with ACS undergoing primary percutaneous coronary intervention under near-infrared spectroscopy intravascular ultrasound (NIRS-IVUS) guidance were included in a prospective manner. Patients were divided into two groups according to the median of the GRACE risk score. Coronary lipid plaques in the target vessel were assessed by NIRS-IVUS with lipid core burden index (LCBI) and a maximum LCBI in 4 mm (maxLCBI4mm). The receiver operating characteristics (ROC) curve analysis was performed based on the major adverse cardiovascular events as an exploratory analysis. The GRACE risk score was significantly and positively correlated with LCBI (r = 0.31, p = 0.03) and maxLCBI4mm (r = 0.38, p = 0.006). LCBI (111.7 ± 85.7 vs. 169.0 ± 83.5, p = 0.02) and maxLCBI4mm (428.5 ± 227.1 vs. 600.6 ± 227.7, p = 0.009) in the target vessel were significantly higher in the high GRACE risk score group than their counterpart. In the ROC curve analysis, LCBI and maxLCBI4mm were predictive for clinical events. In conclusion, the higher GRACE risk score may serve as a discriminator of risk comprising more lipid-rich plaques as an underlying mechanism of an increased risk of thrombotic events after ACS. In patients with ACS, the higher GRACE risk score was significantly and modestly associated with greater coronary lipid plaques in the target vessel.

Discordance between fractional flow reserve and instantaneous wave-free ratio in patients with severe aortic stenosis: A retrospective cohort study

BACKGROUND

Discordance between fractional flow reserve (FFR) and instantaneous wave-free ratio (iFR) occurs in approximately 20 % of cases. However, no studies have reported the discordance in patients with severe aortic stenosis (AS). We aimed to evaluate the diagnostic discordance between FFR and iFR in patients with severe AS.

METHODS

We examined 140 consecutive patients with severe AS (164 intermediate coronary artery stenosis vessels). FFR and iFR were calculated in four quadrants based on threshold FFR and iFR values of ≤0.8 and ≤0.89, respectively (Group 1: iFR >0.89, FFR >0.80; Group 2: iFR ≤0.89, FFR >0.80; Group 3: iFR >0.89, FFR ≤0.80; and Group 4: iFR ≤0.89, FFR ≤0.80). Concordant groups were Groups 1 and 4, and discordant groups were Groups 2 and 3. Positive and negative discordant groups were Groups 3 and 2, respectively.

RESULTS

The median (Q1, Q3) FFR and iFR were 0.84 (0.76, 0.88) and 0.85 (0.76, 0.91), respectively. Discordance was observed in 48 vessels (29.3 %). In the discordant group, negative discordance (Group 2: iFR ≤0.89 and FFR >0.80) was predominant (45 cases, 93.6 %). Multivariate analysis showed that the left anterior descending artery [odds ratio (OR), 3.88; 95 % confidence interval (CI): 1.54-9.79, p = 0.004] and peak velocity ≥5.0 m/s (OR, 3.21; 95%CI: 1.36-7.57, p = 0.008) were independently associated with negative discordance (FFR >0.8 and iFR ≤0.89).

CONCLUSIONS

In patients with severe AS, discordance between FFR and iFR was predominantly negative and observed in 29.3 % of vessels. The left anterior descending artery and peak velocity ≥5.0 m/s were independently associated with negative discordance.

Impact of myocardial bridge on non-culprit vessel lumen changes in patients with acute coronary syndrome

This study aims to clarify the impact of myocardial bridge (MB) on the presence and progression of atherosclerosis in left descending coronary artery (LAD) in patients with acute coronary syndrome (ACS). Ninety-eight patients who underwent percutaneous coronary intervention with the diagnosis of ACS and follow-up coronary angiography but had no significant stenosis in the LAD were included. MB was defined based on coronary angiography. Quantitative coronary angiography was performed to determine the segments where MB was present and proximal to the MB (proximal segment) in patients with MB. In patients without MB, a corresponding region was quantitatively analyzed. The primary endpoint was changes in minimum lumen diameter (MLD) and percentage of diameter stenosis (%DS) in the proximal segment from baseline to follow-up angiography, namely ΔMLD and Δ%DS. MB was identified in 29 (29.6%) patients. Patients with MB had larger MLD and smaller %DS in the proximal segment than their counterpart. During the mean follow-up period of 12.9 ± 5.7 months, MLD and %DS in the proximal segment did not change significantly from baseline to follow-up in patients with and without MB. No significant between-group differences were observed in ΔMLD and Δ%DS. Baseline MLD was identified as the only factor associated with ΔMLD in the proximal segment. ACS patients who had MB but no significant stenosis in the LAD had larger MLD and smaller %DS at the segment proximal to MB compared to those without. In this selected population, serial lumen changes assessed by ΔMLD were not associated with the presence of MB in the LAD.

Risk Factors for Vulnerable Plaque Detected Using Near-Infrared Spectroscopy in Patients Receiving Statin Therapy with No History of Coronary Artery Disease

Residual risk of atherosclerosis remains high despite the use of lipid-lowering therapy with statins. Near-infrared spectroscopy intravascular ultrasound imaging (NIRS-IVUS) can identify vulnerable plaque via the detection of lipid-rich plaque. This study aimed to reveal the clinical characteristics of patients with vulnerable plaque despite statin therapy.NIRS-IVUS was used to determine the maximum 4 mm Lipid Core Burden Index (MaxLCBI_{4 mm}) values of 38 de novo culprit lesions from 32 patients with acute coronary syndrome (53%) (mean age: 73.1 ± 13.1 years) who underwent percutaneous coronary intervention after a minimum 6 months of statin therapy for primary prevention. A patient with vulnerable plaque was defined as an individual presenting at least 1 target lesion with a vulnerable plaque (MaxLCBI_{4 mm} > 400). Overall, the average low-density lipoprotein cholesterol (LDL-C) level was 95.5 ± 27.2 mg/dL. Patients in the vulnerable plaque group were younger and had higher LDL-C, triglycerides, and non-high-density lipoprotein cholesterol (HDL-C) levels than those in the non-vulnerable plaque group. The MaxLCBI_{4 mm} was positively correlated with LDL-C (P = 0.0002), triglycerides (P = 0.0003), and non-HDL-C (P = 0.0001). In multivariate analysis, all 3 treatable lipid components failed to show an independent relationship with the patients with vulnerable plaque. Using receiver-operating characteristics curve analysis, the cutoff points for LDL-C, triglycerides, and non-HDL-C were determined to be 78 mg/dL, 108 mg/dL, and 111 mg/dL, respectively, at MaxLCBI_{4 mm} > 400. In conclusion, this study supports a more comprehensive and aggressive lipid-lowering therapy for the primary prevention of coronary artery disease.

Vascular healing in high-bleeding-risk patients at 3-month after everolimus-eluting stent versus biolimus A9-coated stent implantation: insights from analysis of optical coherence tomography and coronary angioscopy

The prevalence of high-bleeding-risk (HBR) patients who undergo coronary stenting has been reported as 20-40%. This study sought to assess vascular healing in HBR patients by coronary angioscopy (CAS) and optical coherence tomography (OCT). We prospectively analyzed 38 HBR patients with coronary artery disease who successfully underwent everolimus-eluting stent (EES) implantation (20 patients, 23 lesions) or drug-coated stent (DCS) implantation (18 patients, 18 lesions). Follow-up coronary angiography, CAS, and OCT were planned at 3 months after the procedure. The clinical characteristics and inclusion criteria of HBR were comparable between groups. CAS analysis showed that mean yellow color grade was significantly higher with EES than with DCS (1.33 [1.0, 1.67] vs. 1.0 [0.67, 1.5]; P = 0.04). In contrast, OCT analysis demonstrated that most struts in both groups were well-apposed struts with neointimal coverage (93.9% each; P = 1.00), and percentages of the mean neointimal area were comparable between EES and DCS (4.4 ± 3.5 mm² vs. 4.5 ± 4.1 mm²; P = 0.91). The frequency of uncovered struts was significantly lower with EES than with DCS (2.4% vs. 5.3%; P < 0.001), whereas the frequency of malapposed struts was significantly higher with EES than with DCS (3.5% vs. 0.8%; P < 0.001). During follow-up, no stent thrombosis or major bleeding complications were encountered in either group. Among HBR patients, both EES and DCS demonstrated good vascular healing at 3-month follow-up with some different features in CAS and OCT assessments.

Relationship Between Optical Coherence Tomography-Derived In-Stent Neoatherosclerosis and the Extent of Lipid-Rich Neointima by Near-Infrared Spectroscopy and Intravascular Ultrasound: A Multimodal Imaging Study

Background In-stent restenosis, especially for neoatherosclerosis, is a major concern following percutaneous coronary intervention. This study aimed to elucidate the association of features of in-stent restenosis lesions revealed by optical coherence tomography (OCT)/optical frequency domain imaging (OFDI) and the extent of lipid-rich neointima (LRN) assessed by near-infrared spectroscopy (NIRS) and intravascular ultrasound, especially for neoatherosclerosis. Methods and Results We analyzed patients undergoing percutaneous coronary intervention for in-stent restenosis lesions using both OCT/OFDI and NIRS-intravascular ultrasound. OCT/OFDI-derived neoatherosclerosis was defined as lipid neointima. The existence of large LRN (defined as a long segment with 4-mm maximum lipid core burden index ≥400) was evaluated by NIRS. In 59 patients with 64 lesions, neoatherosclerosis and large LRN were observed in 17 (26.6%) and 21 lesions (32.8%), respectively. Naturally, large LRN showed higher 4-mm maximum lipid core burden index (median [interquartile range], 623 [518-805] versus 176 [0-524]; P<0.001). In OCT/OFDI findings, large LRN displayed lower minimal lumen area (0.9±0.4 versus 1.3±0.6 mm²; P=0.02) and greater max lipid arc (median [interquartile range], 272° [220°-360°] versus 193° [132°-247°]; P=0.004). In the receiver operating characteristic curve analysis, 4-mm maximum lipid core burden index was the best predictor for neoatherosclerosis, with a cutoff value of 405 (area under curve, 0.92 [95% CI, 0.83-1.00]). In multivariable logistic analysis, only low-density lipoprotein cholesterol (odds ratio, 1.52 [95% CI, 1.11-2.08]) was an independent predictor for large LRNs. Conclusions NIRS-derived large LRN was significantly associated with neoatherosclerosis by OCT/OFDI. The neointimal characterization by NIRS-intravascular ultrasound has potential as an alternative method of OCT/OFDI for in-stent restenosis lesions.

Long-term outcomes of intermediate coronary stenosis in patients undergoing hemodialysis after deferred revascularization based on fractional flow reserve

OBJECTIVES

This study aimed to assess the long-term outcomes of patients undergoing hemodialysis (HD) after deferred revascularization based on fractional flow reserve (FFR).

BACKGROUND

FFR is a practical technique for assessing the functional severity of intermediate coronary stenosis. Prior research has revealed a satisfactory outcome in patients after the deferral of percutaneous coronary intervention for coronary lesions based on FFR measurement. However, little research has been conducted focusing on patients undergoing HD.

METHODS

The retrospective study comprised 225 consecutive patients with FFR assessment and deferred revascularization between January 2016 and December 2019. Based on a deferral cutoff FFR value of >0.80, we assessed the differences in all-cause death, major adverse cardiac events (MACEs), and target vessel failure (TVF) between the HD (n = 69) and non-HD groups (n = 156) during a mean ± standard deviation routine follow-up of 32.2 ± 13.4 months.

RESULTS

Although the HD group had significantly higher rates of diabetes mellitus than the non-HD group (53.6% vs. 37.2%, p = 0.021), there were no significant differences in sex, left ventricular ejection fraction, or other risk factors between the groups, nor with respect to stenosis diameter or mean FFR. The HD group had a significantly higher incidence of TVF than the non-HD group (34.8% vs. 14.1%, p < 0.001), as well as a significantly higher risk of all-cause death and MACEs.

CONCLUSIONS

The study revealed that deferred revascularization in coronary lesions with an FFR value of >0.80 in patients undergoing HD was associated with poor outcomes. Therefore, it is important to carefully monitor patients with intermediate coronary stenosis undergoing HD.

Effect of Stenting Strategy on the Outcome in Patients with Non-Left Main Bifurcation Lesions

Previous studies have not compared outcomes between different percutaneous coronary intervention (PCI) strategies and lesion locations in non-left main (LM) bifurcation lesions. We enrolled 2044 patients from a multicenter registry with an LAD bifurcation lesion (n = 1551) or non-LAD bifurcation lesion (n = 493). The primary outcome was target lesion failure (TLF), a composite of cardiac death, myocardial infarction, and target lesion revascularization (TLR). During a median follow-up period of 38 months, non-LAD bifurcation lesions treated with the two-stent strategy, compared with the one-stent strategy, were associated with more frequent TLF (20.7% vs. 6.3%, p < 0.01), TLR (16.7% vs. 4.7%, p < 0.01), and target vessel revascularization (TVR; 18.2% vs. 6.3%, p < 0.01). There was no significant difference in outcome among LAD bifurcation lesions treated with different PCI strategies. The two-stent strategy was associated with a higher risk of TLF (adjusted HR 4.34, CI 1.93−9.76, p < 0.01), TLR (adjusted HR 4.30, CI 1.64−11.27, p < 0.01), and TVR (adjusted HR 5.07, CI 1.69−9.74, p < 0.01) in the non-LAD bifurcation lesions. The planned one-stent strategy is preferable to the two-stent strategy for the treatment of non-LAD bifurcation lesions.

Use of a Cutting Balloon Reduces the Incidence of Distal Embolism in Acute Coronary Syndrome Requiring Predilatation Before Stenting

Background: Acute coronary syndrome (ACS) patients with solid lesions often require predilatation before stenting. Predilatation with high pressure may increase the risk of distal embolism, whereas direct stenting increases the risk of stent underexpansion. We recently reported that, in severely calcified lesions, using a cutting balloon (CB) can provide greater acute gain compared with other scoring balloons. Therefore, we hypothesized that predilatation with CB may reduce the incidence of distal embolism in ACS patients with solid lesions.

Methods and Results: This study retrospectively analyzed data for 175 ACS patients who required predilatation, either with a conventional balloon (n=136) or CB (n=39). The occurrence of distal embolism was significantly lower in the CB than conventional balloon group (10.3% vs 32.4%, respectively; P=0.007). Multivariate analysis showed that the occurrence of distal embolism was positively associated with Thrombolysis in Myocardial Infarction (TIMI) grade and the presence of attenuated plaque, but negatively associated with the use of a CB. To support this clinical observation, we compared thrombus dispersal using a CB and non-compliant balloon in an ex vivo experimental model using a pseudo-thrombus. In this model, pseudo-thrombus dispersal was significantly smaller when a CB rather than non-compliant balloon was used (1.8±1.0% vs 2.6±1.2%, respectively; n=20, for each; P=0.002).

Conclusions: In ACS patients with solid lesions that require predilatation, predilatation with a CB may reduce the incidence of distal embolism.

Relation of hyperemic response during fractional flow reserve measurement to coronary flow reserve

BACKGROUND

Coronary flow reserve (CFR) represents entire coronary compensatory capacity, while fractional flow reserve (FFR) is a standard to evaluate functional severity of epicardial coronary artery disease (CAD). ΔFFR, a decrease in a ratio of mean distal coronary pressure to aortic pressure (Pd/Pa) at rest to FFR, is conceptually associated with coronary microvascular function. This study aimed to evaluate the relation of ΔFFR to CFR in patients with stable CAD.

METHODS

We performed resting Pd/Pa and FFR measurements in a total of 309 vessels with intermediate coronary artery stenosis in 242 patients. ΔFFR was defined as (resting Pd/Pa - FFR), and pressure bounded-CFR was calculated to estimate low CFR. Vessels were divided as the low CFR and non-low CFR groups.

RESULTS

Of 309 vessels, low CFR was observed in 101 (32.7 %). While FFR values were similar (0.78 ± 0.11 vs. 0.78 ± 0.09, p = 0.84), resting Pd/Pa (0.85 ± 0.08 vs. 0.93 ± 0.04, p < 0.001) and ΔFFR (0.07 ± 0.06 vs. 0.15 ± 0.06, p < 0.001) were significantly lower in the low CFR group than in the non-low CFR group. The receiver operating characteristic curve analysis indicated that ΔFFR was predictive for low CFR (area under the curve 0.84, best cut-off value 0.08, p < 0.001). Multivariable analysis identified lower ΔFFR, the left anterior descending coronary artery, and lower hemoglobin and higher brain natriuretic peptide levels as factors associated with low CFR.

CONCLUSIONS

In patients with stable CAD, lower ΔFFR was significantly associated with low CFR in intermediate coronary stenosis in patients with stable CAD. ΔFFR may be a simple, practical, and useful surrogate to identify patients with impaired CFR.

The Residual Lipid-rich Coronary Atheroma behind the Implanted Newer-generation Drug-eluting Stent and Future Stent-related Event's Risks

BACKGROUND

Lipid-rich plaque is an important substrate that causes future coronary events. However, the clinical implications of underlying plaque characteristics in coronary lesions after newer-generation drug-eluting stent (DES) implantation remain unknown.

METHODS

The current study analyzed 445 target lesions after newer-generation DES implantation in 416 patients with CAD (CCS/ACS=264/181) from the REASSURE-NIRS multi-center registry. Near-infrared spectroscopy (NIRS) imaging was used to evaluate maximum lipid core burden index after stent implantation in target lesions (residual maxLCBI_4mm). The primary and secondary outcomes were 3-year lesion-oriented clinical outcomes (LOCO; cardiac death, non-fatal target lesion-related MI, or ischemia-driven TLR) and patient-oriented clinical outcomes (POCO; all-cause death, non-fatal MI, or ischemia-driven unplanned revascularization). Outcomes were compared by residual maxLCBI_4mm tertile.

RESULTS

Median residual maxLCBI_4mm was 183; 16% of lesions had residual maxLCBI_4mm >400. Higher residual maxLCBI_4mm was not associated with a greater likelihood of LOCO or POCO during the observational period (LOCO, log-rank p=0.76; POCO, log-rank p=0.84). Mixed-effects logistic regression demonstrated that residual maxLCBI_4mm does not predict LOCO (odds ratio=1.000, 95% CI=0.997-1.003, p=0.95). There was no significant relationship between residual maxLCBI_4mm and POCO (odds ratio=1.001, 95% CI=0.999-1.002, p=0.30).

CONCLUSIONS

Residual maxLCBI_4mm is not associated with LOCO or POCO in patients with CAD after newer-generation DES implantation. Our findings suggest that NIRS-derived underlying lipid-rich plaque is not associated with the risk of stent-related events and patient-based outcomes in patients with CAD who have received a newer-generation DES.

Title: how accurately can machine learning technology for quantitative imaging analysis represent actual coronary artery data?

The development of quantitative imaging analysis of the human coronary artery is an important contribution to daily clinical practice and research in cardiovascular medicine. However, cardiovascular data that are directly acquired are not identical to data acquired using imaging analysis. Whereas the accuracy and reproducibility of machine learning (ML) analysis on the lumen, vessel, and plaque areas are developed, many diverse efforts are still needed to provide the best patient care.

Comparison of the Incidence of Periprocedural Myocardial Infarction in Bifurcation Lesions Between Medina (1,1,1) and (0,1,1) in Elective Percutaneous Coronary Intervention

Periprocedural myocardial infarction (PMI) following percutaneous coronary intervention (PCI) is more frequently observed in true bifurcation lesions such as Medina (1,1,1) and (0,1,1). The aim of this study is to compare the incidence of PMI in elective PCI between Medina (1,1,1) and (0,1,1) bifurcation lesions. This was a retrospective, single-center study. We included 162 true bifurcation lesions, which were divided into the (1,1,1) group (n = 85) and the (0,1,1) group (n = 77). We compared the incidence of PMI between the two groups and performed multivariate logistic regression analysis using PMI as a dependent variable. The incidence of PMI was similar in the (1,1,1) group and the (0,1,1) group (12.9% versus 15.6%, P = 0.658). The final TIMI flow grade of the side branches and that of the main branches were also similar in the two groups. In multivariate logistic regression analysis, Medina classification (1,1,1) was not associated with PMI (odds ratio (OR), 0.996; 95% confidence interval (CI), 0.379-2.621; P = 0.994), but the angle of the side branch < 45° (OR, 3.569; 95% CI, 1.320-9.654; P = 0.012), lesion length in a main vessel (per 10-mm increase) (OR, 1.508; 95% CI, 1.104-2.060; P = 0.010), and absence of side branch protection (OR, 3.034; 95% CI, 1.095-8.409; P = 0.033) were significantly associated with PMI. In conclusion, the Medina (1,1,1) bifurcation lesions did not increase the incidence of PMI as compared to Medina (0,1,1). However, the narrow side branch angle, diffuse long lesion, and absence of side branch protection were significantly associated with PMI. We should pay attention to these high-risk features in the treatment of true bifurcation lesions.

Modern atlas of invasive coronary angiography views: a practical approach for fellows and young interventionalists

Although computed tomographic (CT) assessment of coronary arteries is rapidly becoming a first line imaging technique for the assessment of coronary artery disease (CAD), coronary angiography (CA) still represents the gold standard method, especially for guiding percutaneous coronary artery interventions (PCI). Together with the basic principles of radioprotection and modern radiological equipment's management, the average invasive cardiology should know how to replicate in clinical practice, both the old standard and the new and previously under-used projections and should need to learn their usefulness at both diagnostic and interventional levels. The present article is aimed to present an updated practical overview of modern available angiographic views for fellows and young interventionalists.

EROSION III: A Multicenter RCT of OCT-Guided Reperfusion in STEMI With Early Infarct Artery Patency

OBJECTIVES

The aim of this study was to test whether optical coherence tomographic (OCT) guidance would provide additional useful information beyond that obtained by angiography and lead to a shift in reperfusion strategy and improved clinical outcomes in patients with ST-segment elevation myocardial infarction (STEMI) with early infarct artery patency.

BACKGROUND

Angiography is limited in assessing the underlying pathophysiological mechanisms of the culprit lesion.

METHODS

EROSION III (Optical Coherence Tomography-Guided Reperfusion in ST-Segment Elevation Myocardial Infarction With Early Infarct Artery Patency) is an open-label, prospective, multicenter, randomized, controlled study approved by the ethics committees of participating centers. Patients with STEMI who had angiographic diameter stenosis ≤ 70% and TIMI (Thrombolysis In Myocardial Infarction) flow grade 3 at presentation or after antegrade blood flow restoration were recruited and randomized to either OCT guidance or angiographic guidance. The primary efficacy endpoint was the rate of stent implantation.

RESULTS

Among 246 randomized patients, 226 (91.9%) constituted the per protocol set (112 with OCT guidance and 114 with angiographic guidance). The median diameter stenosis was 54.0% (IQR: 48.0%-61.0%) in the OCT guidance group and 53.5% (IQR: 43.8%-64.0%) in the angiographic guidance group (P = 0.57) before randomization. Stent implantation was performed in 49 of 112 patients (43.8%) in the OCT group and 67 of 114 patients (58.8%) in the angiographic group (P = 0.024), demonstrating a 15% reduction in stent implantation with OCT guidance. In patients treated with stent implantation, OCT guidance was associated with a favorable result with lower residual angiographic diameter stenosis (8.7% ± 3.7% vs 11.8% ± 4.6% in the angiographic guidance group; P < 0.001). Two patients (1 cardiac death, 1 stable angina) met the primary safety endpoint in the OCT guidance group, as did 3 patients (3 cardiac deaths) in the angiographic guidance group (1.8% vs 2.6%; P = 0.67). Reinfarction was not observed in either group. At 1 year, the rates of predefined cardiocerebrovascular events were comparable between the groups (11.6% after OCT guidance vs 9.6% after angiographic guidance; P = 0.66).

CONCLUSIONS

In patients with STEMI with early infarct artery patency, OCT guidance compared with angiographic guidance of reperfusion was associated with less stent implantation during primary percutaneous coronary intervention. These favorable results indicate the value of OCT imaging in optimizing the reperfusion strategy of patients with STEMI. (EROSION III: OCT- vs Angio-Based Reperfusion Strategy for STEMI; NCT03571269).

Long-Term Outcomes and Clinical Predictors of Mortality Following Occurrence of Stent Thrombosis

Background Stent thrombosis (ST) remains a significant medical issue. In particular, longer-term mortality and clinical predictors after ST occurrence have yet to be elucidated. Methods and Results This was a multicenter, retrospective, observational study. A total of 187 definite ST cases from January 2008 to December 2017 were enrolled, and the long-term clinical outcomes were investigated. The primary outcome measure was the cumulative mortality after ST occurrence. In addition, independent predictors of mortality were assessed. Among the stent types causing ST, bare-metal stent, first-generation drug-eluting stent, second-generation drug-eluting stent, and third-generation drug-eluting stent comprised 31.0%, 19.3%, 36.9%, and 6.4% of cases, respectively. Median duration from stent implantation to ST was 680.5 (interquartile range, 33.8-2450.5) days. Cumulative mortality was 14.6%, 17.4%, 21.2%, 24.4%, and 33.8% at 1, 2, 3, 5 and 10 years, respectively. The cumulative mortality did not significantly differ by type of stent, and mortality of late ST was higher than that of early ST and very late ST; however, it did not reach statistical significance after the multivariate analysis. Independent predictors of mortality were hemodialysis (hazard ratio [HR], 7.80; 95% CI, 3.07-19.81; P<0.001), culprit lesions in the left main trunk (HR, 8.14; 95% CI, 1.71-38.75; P=0.008), culprit lesions in the left coronary artery (HR, 2.77; 95% CI, 1.10-6.96; P=0.030), and peak creatine kinase (HR, 1.017; 95% CI, 1.011-1.022; P<0.001). Conclusions The 10-year cumulative mortality after ST reached 33.8%. Close follow-up is thus mandatory for patients with ST, especially with hemodialysis, culprit lesions in the left main trunk and left coronary artery, and high peak creatine kinase.

Relation of glucose variability to vulnerable plaque formation in patients with coronary artery disease

BACKGROUND

Although glucose variability (GV) is reportedly associated with coronary plaque vulnerability, namely lipid-rich plaque, details are not fully understood. The aim of this study was to evaluate relations of GV after discharge to vulnerable plaque formation assessed by near-infrared spectroscopy intravascular ultrasound (NIRS-IVUS) in patients with and without diabetes.

METHODS

A total of 40 patients undergoing percutaneous coronary intervention under NIRS-IVUS guidance were included, among whom 13 (33%) had diabetes and 20 (50%) presented with acute myocardial infarction (MI). GV was evaluated by a flush glucose monitoring system, primarily with mean amplitude of glycemic excursion (MAGE). Lipid-rich plaque was assessed by maximum lipid core burden index in 4 mm (maxLCBI_4mm) in the target lesion using NIRS-IVUS.

RESULTS

Mean MAGE and maxLCBI_4mm were 69.7 ± 25.6 mg/dl and 508.0 ± 294.9. Intra-day GV was not significantly associated with maxLCBI_4mm in the entire study population, while MAGE was correlated with maxLCBI_4mm in non-diabetic patients (r = 0.46, p = 0.02). In patients with and without acute MI presentation, no significant relations were found between MAGE and maxLCBI_4mm.

CONCLUSION

GV was associated with lipid core plaque formation, especially in non-diabetic patients.

Discrepancy between plaque vulnerability and functional severity of angiographically intermediate coronary artery lesions

This study sought to investigate the relationship between physiological severity and plaque vulnerability of intermediate coronary artery stenoses as assessed by fractional flow reserve (FFR) and near-infrared spectroscopy-intravascular ultrasound (NIRS-IVUS). We included vessels where both FFR and NIRS-IVUS were performed. A positive FFR was defined as FFR ≤ 0.80. Lipid core burden index of the entire target vessel (TV-LCBI), maximum LCBI in 4 mm (maxLCBI_4mm), and maximum plaque burden (PB) were evaluated using NIRS-IVUS. A vulnerable plaque was defined as a lipid-rich plaque (maxLCBI4_mm ≥ 400) with large PB (≥ 70%). A total of 59 vessels of 45 patients were included. Median FFR value was 0.75 [interquartile 0.72, 0.82]. An FFR value of ≤ 0.80 was observed in 42 vessels (71%). TV-LCBI (correlation coefficient [CC] = - 0.331, p = 0.011), lesion length (CC = - 0.350, p = 0.007), and PB (CC = - 0.230, p = 0.080) negatively correlated with FFR value, while maxLCBI_4mm did not (CC = - 0.156, p = 0.24). The prevalence of vulnerable plaques (26.2% vs. 29.4%, p > 0.99) and mean TV-LCBI, maxLCBI_4mm, and PB values were not significantly different between the vessels with FFR ≤ 0.80 and those with FFR > 0.80. In multivariable logistic models, diabetes mellitus (p = 0.003) and hemoglobin A1c (p = 0.012) were associated with the presence of a vulnerable plaque. In conclusion, the results of the present study suggested that FFR may reflect total lipid burden but not necessarily plaque vulnerability. In patients with coronary artery disease and a high likelihood of rapid atherosclerosis progression, such as diabetes mellitus patients, assessing plaque vulnerability in addition to the functional severity of coronary artery lesions may help stratify better the risk of future events.

Factors associated with difficulty in crossing the culprit lesion of acute myocardial infarction

In percutaneous coronary intervention (PCI) to the culprit lesion of acute myocardial infarction (AMI), unsuccessful guidewire crossing causes immediate poor outcomes. It is important to determine the factors associated with unsuccessful guidewire crossing in AMI lesions. The purpose of this study was to find factors associated with difficulty in crossing the culprit lesion of AMI. We defined the difficult group when the guidewire used to cross the culprit lesion was a polymer jacket type guidewire or a stiff guidewire. We included 937 patients, and divided those into the non-difficult group (n = 876) and the difficult group (n = 61). Proximal reference diameter was significantly smaller in the difficult group than in the non-difficult group (p < 0.001), and degree of calcification was severer in the difficult group than in the non-difficult group (p < 0.001). In the multivariate stepwise logistic regression analysis, proximal reference diameter [odds ratio (OR) 0.313, 95% confidence interval (CI) 0.185–0.529, p < 0.001)], previous PCI (OR 3.065, 95% CI 1.612–5.830, p = 0.001), moderate-severe calcification (OR 4.322, 95% CI 2.354–7.935, p < 0.001), blunt type obstruction (OR 12.646, 95% CI 6.805–23.503, p < 0.001), and the presence of collateral to the culprit lesion (OR 2.110, 95% CI 1.145–3.888, p = 0.017) were significantly associated with difficulty in crossing the culprit lesion. In conclusion, previous PCI, calcification, blunt type obstruction, and the presence of collateral were associated with difficulty in crossing the culprit lesion, whereas proximal reference diameter was inversely associated with difficulty. Our study provides a reference to recognize the difficulty in crossing the culprit lesions of AMI for PCI operators, especially junior operators.

Proximal optimisation technique versus final kissing balloon inflation in coronary bifurcation lesions: the randomised, multicentre PROPOT trial

BACKGROUND

Clinical implications of the proximal optimisation technique (POT) for bifurcation lesions have not been investigated in a randomised controlled trial.

AIMS

This study aimed to investigate whether POT is superior in terms of stent apposition compared with the conventional kissing balloon technique (KBT) in real-life bifurcation lesions using optical coherence tomography (OCT).

METHODS

A total of 120 patients from 15 centres were randomised into two groups - POT followed by side branch dilation or KBT. Finally, 57 and 58 patients in the POT and KBT groups, respectively, were analysed. OCT was performed at baseline, immediately after wire recrossing to the side branch, and at the final procedure.

RESULTS

The primary endpoint was the rate of malapposed struts assessed by the final OCT. The rate of malapposed struts did not differ between the POT and KBT groups (in-stent proximal site: 10.4% vs 7.7%, p=0.33; bifurcation core: 1.4% vs 1.1%, p=0.67; core's distal edge: 6.2% vs 5.3%, p=0.59). More additional treatments were required among the POT group (40.4% vs 6.9%, p<0.01). At one-year follow-up, only one patient in each group underwent target lesion revascularisation (2.0% vs 1.9%).

CONCLUSIONS

POT followed by side branch dilation did not show any advantages over conventional KBT in terms of stent apposition; however, excellent midterm clinical outcomes were observed in both strategies.

Clinical Efficacy of Intracoronary Papaverine After Nicorandil Administration for Safe and Optimal Fractional Flow Reserve Measurement

Fractional flow reserve (FFR) is considered the standard for assessment of the physiological significance of coronary artery stenosis. Intracoronary papaverine (PAP) is the most potent vasodilator used for the achievement of maximal hyperemia. However, its use can provoke ventricular tachycardia (VT) due to excessive QT prolongation. We evaluated the clinical efficacy and safety of the administration of PAP after nicorandil (NIC), a potassium channel opener that prevents VT, for optimal FFR measurement.A total of 127 patients with 178 stenoses were enrolled. The FFR values were measured using NIC (NIC-FFR) and PAP (PAP-FFR). We administered PAP following NIC (NIC-PAP). Changes in the FFR and electrogram parameters (baseline versus NIC versus PAP) were assessed and the incidence of arrhythmias after PAP was evaluated. In addition, we analyzed another 41 patients with 51 stenoses by assessing the FFR using PAP before NIC (PAP-NIC). After propensity score matching, the electrogram parameters between 2 groups were compared.The mean PAP-FFR was significantly lower than the mean NIC-FFR (0.82 ± 0.11 versus 0.81 ± 0.11, P < 0.05). The mean baseline-QTc, NIC-QTc, and PAP-QTc values were 425 ± 37 ms^1/2, 424 ± 41 ms^1/2, and 483 ± 54 ms^1/2, respectively. VT occurred in only 1 patient (0.6%). Although PAP induced QTc prolongation (P < 0.05), the PAP-QTc duration was significantly shorter in NIC-PAP compared to PAP-NIC (P < 0.05).The administration of PAP with NIC may induce sufficient hyperemia and prevent fatal arrhythmia through reductions in the PAP-induced QTc prolongation during FFR measurement.

Stent visualization methods to guide percutaneous coronary interventions and assess long-term patency

Evaluation of acute percutaneous coronary intervention (PCI) results and long-term follow-up remains challenging with ongoing stent designs. Several imaging tools have been developed to assess native vessel atherosclerosis and stent expansion, improving overall PCI results and reducing adverse cardiac events. Quantitative coronary analysis has played a crucial role in quantifying the extent of coronary artery disease and stent results. Digital stent enhancement methods have been well validated and improved stent strut visualization. Intravascular imaging remains the gold standard in PCI guidance but adds costs and time to the procedure. With a recent shift towards non-invasive imaging assessment and coronary computed tomography angiography imaging have shown promising results. We hereby review novel stent visualization techniques used to guide PCI and assess stent patency in the modern PCI era.

Research on the Method of Predicting Fractional Flow Reserve Based on Multiple Independent Risk Factors

The use of diameter stenosis (DS), as revealed by coronary angiography, for predicting fractional flow reserve (FFR) usually results in a high error rate of detection. In this study, we investigated a method for predicting FFR in patients with coronary stenosis based on multiple independent risk factors. The aim of the study was to improve the accuracy of detection. First, we searched the existing literature to identify multiple independent risk factors and then calculated the corresponding odds ratios. The improved analytic hierarchy process (IAHP) was then used to determine the weighted value of each independent risk factor, based on the corresponding odds ratio. Next, we developed a novel method, based on the top seven independent risk factors with the highest weighted values, to predict FFR. This model was then used to predict the FFR of 253 patients with coronary stenosis, and the results were then compared with previous methods (DS alone and a simplified scoring system). In addition to DS, we identified a range of other independent risk factors, with the highest weighted values, for predicting FFR, including gender, body mass index, location of stenosis, type of coronary artery distribution, left ventricular ejection fraction, and left myocardial mass. The area under the receiver-operating characteristic curve (AUC) for the newly developed method was 84.3% (95% CI: 79.2–89.4%), which was larger than 65.3% (95% CI: 61.5–69.1%) of DS alone and 74.8% (95% CI: 68.4–81.2%) of the existing simplified scoring system. The newly developed method, based on multiple independent risk factors, effectively improves the prediction accuracy for FFR.

Impact of glycemic variability on coronary and peripheral endothelial dysfunction in patients with coronary artery disease

BACKGROUND

Previous studies have reported that glucose variability leads to endothelial dysfunction and progression of coronary atherosclerosis. However, few studies have directly evaluated the relation between glucose variability and coronary endothelial function in patients with coronary artery disease (CAD).

METHODS

A total of 38 patients with chronic CAD and a history of coronary drug-eluting stent implantation were enroled. Coronary endothelial function was evaluated by measuring the coronary vasoreactivity using quantitative coronary angiography in the segment distal to implanted stent in response to intracoronary acetylcholine (ACh) infusion (10-7 mol/l). Peripheral endothelial function was also assessed with reactive hyperemia index (RHI). The mean amplitude of glycemic excursion (MAGE) was calculated as a primary metric of glucose variability using a flash glucose monitoring system.

RESULTS

Of 38 patients, 17 (45%) had diabetes mellitus. The mean levels of glycated hemoglobin, MAGE, and RHI were 6.3 ± 0.8%, 71.4 ± 29.8 mg/dl, and 1.85 ± 0.63. In the distal segment to coronary stent, lumen diameter was constricted by 0.6 ± 7.3% in response to intracoronary ACh infusion compared to that at baseline. While peripheral endothelial function assessed with RHI was not significantly associated with MAGE (r = -0.16, p = 0.35), coronary endothelial function was correlated with MAGE (r = -0.38, p = 0.02).

CONCLUSION

Greater glucose variability was significantly associated with coronary rather than peripheral endothelial dysfunction in patients with CAD, suggesting an impact of glucose variability on coronary atherosclerosis.

Association between non-culprit healed plaque and plaque progression in acute coronary syndrome patients: an optical coherence tomography study

BACKGROUND

Healed plaques are frequently found in patients with acute coronary syndrome, but the prognostic value is debatable. This study investigated the clinical features of non-culprit healed plaques detected by optical coherence tomography (OCT) with the aim of predicting plaque progression of healed plaques.

METHODS

This study retrospectively analyzed 113 non-culprit lesions from 85 patients who underwent baseline OCT imaging and follow-up angiography from January 2015 to December 2019. Plaque progression predictors were assessed by multivariate analysis.

RESULTS

Among 113 non-culprit lesions, 27 healed plaques (23.9%) were identified. Patients with non-culprit healed plaques had prior antiplatelet therapy (65.0% vs. 33.8%, P = 0.019), hypertension (85.0% vs. 50.7%, P = 0.009), and dyslipidemia (70.0% vs. 41.5%, P = 0.04) which were more frequently than those without healed plaques. The thickness (r = 0.674, P < 0.001), arc ( r = 0.736, P < 0.001), and volume ( r = 0.541, P = 0.004) of healed plaque were correlated with minimum lumen diameter changes. At a mean follow-up of 11.5 months, the non-culprit healed plaques had a lower minimum lumen diameter (1.61 ± 0.46 mm vs. 1.91 ± 0.73 mm, P = 0.016), lower average lumen diameter (1.86 mm vs. 2.10 mm, P = 0.033), and a higher degree of diameter stenosis (41.4% ± 11.9% vs. 35.5% ± 13.1%, P = 0.031) when compared to baseline measurements. The plaque progression rate was higher in the healed plaque group (33.3% vs. 8.1%, P = 0.002), and multivariate analysis identified healed plaques [odds ratio (OR) = 8.49, 95% CI: 1.71-42.13] and lumen thrombus (OR = 10.69, 95% CI: 2.21-51.71) as predictors of subsequent lesion progression.

CONCLUSIONS

Healed plaques were a predictor for rapid plaque progression. The quantitative parameters of healed plaque showed a good agreement with plaque progression. Patients with healed plaque were associated with prior antiplatelet therapy and high level of low-density lipoprotein cholesterol. Bifurcation lesions might be the predilection sites of healed plaques.

Impact of in-stent tissue characteristics on excimer laser coronary angioplasty prior to drug-coated balloon treatment

BACKGROUND

Percutaneous coronary intervention with drug-coated balloon (DCB) angioplasty is one of the standard treatments for lesions with in-stent restenosis (ISR). However, the efficacy of additional excimer laser coronary angioplasty (ELCA) for ISR lesions prior to DCB angioplasty has not been elucidated. The aims of this study were to elucidate the efficacy of ELCA prior to DCB treatment for ISR and the difference in effectiveness by lesion morphology.

METHODS

This was a multicenter, retrospective observational study. We enrolled 208 ISR lesions from 204 patients which were treated with DCB angioplasty under optical coherence tomography (OCT) guidance. We compared the acute gain evaluated by quantitative coronary angiography and clinically driven target lesion revascularization (CD-TLR) between the ELCA (+) (47 lesions) and ELCA (-) (161 lesions) groups.

RESULTS

The acute gain was significantly larger in the ELCA (+) group (1.51 ± 0.47 mm versus 1.29 ± 0.52 mm, P = 0.012). Even after adjustment for comorbidities, the ELCA had a significant impact on the acute gain (coefficient 0.24 [95% confidence interval 0.067-0.41]). In addition, the ELCA usage was significantly associated with larger acute gain in lesions with a homogeneous pattern, although there was no association between ELCA and either the heterogeneous pattern or neoatherosclerosis. The freedom from CD-TLR was not different between the ELCA (-) and ELCA (+) groups (hazard ratio 0.53 [0.24-1.18]).

CONCLUSIONS

ELCA had a significant impact on the larger acute gain in ISR lesions, especially may did in those with a homogenous pattern, while it did not impact CD-TLR significantly.

Effect of Paclitaxel-Coated Balloon Angioplasty on Side Branch Lesion and Cardiovascular Outcomes in Patients with De Novo True Coronary Bifurcation Lesions Undergoing Percutaneous Coronary Intervention

PURPOSE

To compare the effects of paclitaxal-coated balloon (PCB) versus conventional balloon (CB) on side branch (SB) lesion and cardiovascular outcomes in patients with de novo true bifurcation lesions.

METHODS

In total, 219 patients with de novo true bifurcation lesions were enrolled and divided into PCB group (102 cases) and CB group (117 cases) according to angioplasty strategy in SB. Drug-eluting stent (DES) was implanted in main vessel (MV) for each subject. All subjects underwent a 12-month follow-up for late lumen loss (LLL), restenosis, and major adverse cardiovascular events (MACE) after percutaneous coronary intervention (PCI). MACEs included cardiac death, nonfatal myocardial infarction, and unstable angina requiring admission.

RESULTS

There were no differences in diameter, minimum lumen diameter (MLD), and stenosis for bifurcation lesions between the two groups before and immediately after PCI (P > 0.05). After 12-month follow-up, no differences occurred in MV-MLD and MV-LLL between the two groups (P > 0.05); SB-MLD in PCB group was higher than that in CB group (1.97 ± 0.36 mm vs. 1.80 ± 0.43 mm, P = 0.007); SB-LLL in PCB group was lower than that in CB group (0.11 ± 0.18 mm vs. 0.19 ± 0.25 mm, P = 0.024). Multivariate COX analyses indicated that PCB group had lower MACE risk than CB group (HR = 0.480, 95%CI 0.244-0.941, P = 0.033).

CONCLUSION

PCB could decrease SB-LLL and MACE risk in patients with de novo true coronary bifurcation lesion 12 months after single-DES intervention.